International party assistance –
what do we know about the effects?
Lars Svåsand
University of Bergen
Rapport 2014:03
till
Expertgruppen för biståndsanalys (EBA)
Acknowledgements: I thank the members of the reference group: Eva
Lindström, Stefan Eriksson, Johan Norqvist, Helena Bjuremalm,
Fredrik Uggla and Anne Mette Kjær, and the EBA secretariat staff
members Sonja Daltung and Emma Öståker, for valuable comments
on earlier drafts. The opinions expressed in the manuscript and any
errors or mistakes are those of the author.
“Building effective party structures is an endless task. Healthy
organizations can and will adapt to changing circumstances”
National Democratic Institute on twitter (accessed 07.10.14)
This report can be downloaded free of charge at www.eba.se
This work is licensed under the Creative Commons Attribution 4.0 International
License. To view a copy of this license, visit
http://creativecommons.org/licenses/by/4.0/.
ISBN: 978-91-88143-00-6
Printed by Elanders Sverige AB
Stockholm 2014
Cover design by Julia Demchenko
Table of contents
Preface ............................................................................... 1
Sammanfattning .................................................................. 3
Summary ............................................................................ 6
1. Introduction .................................................................... 9
1.1 Democracy assistance and party assistance ............................... 10
1.2 Method, material and limitations ............................................... 11
1.3 Outline ......................................................................................... 12
2.1 The necessity of parties in democracies and the
background for IPA ................................................................. 13
2.2 The challenges of parties and party systems in new
democracies .............................................................................. 16
3. Dimensions of International Party Assistance .................... 23
3.1 Objectives of IPA ....................................................................... 23
3.2
Organizational models for IPA .............................................. 31
3.3 Financial model and scale of intervention ................................. 33
3.4 Modes of intervention ................................................................ 35
3.5 Timing and endurance of IPA projects...................................... 37
3.6 Geographic focus ........................................................................ 38
3.7 Summary ...................................................................................... 40
4. Assessing the effects of IPA ............................................ 42
4.1 Challenges in assessing the effects of IPA ................................ 42
4.2. Does IPA work or not? ............................................................. 46
5. Concluding reflections .................................................... 52
5.1 A new IPA agenda? .................................................................... 52
5.2 Should IPA be maintained?........................................................ 54
5.3 If IPA is maintained: what should be done? ............................. 57
5.4 Key factors to consider............................................................... 59
Appendix A: Findings from research literature on IPA............. 65
Appendix B: IPA donors and findings from evaluations ........... 73
Sweden: support through party-affiliated organizations ............... 74
Denmark: The Danish Institute for Parties and Democracy ......... 81
Norway: The Norwegian Center for Democracy .......................... 83
US institutions .................................................................................. 84
UK: The Westminster Foundation for Democracy ....................... 89
Netherlands Institute for Multiparty Democracy .......................... 91
References ........................................................................ 99
Preface
Sweden has a long tradition of providing development cooperation
with the aim of strengthening democracy and the respect for human
rights. Democratic development has been a thematic priority for many
years, and democracy support constitutes one of the largest parts of
Swedish aid. Given the strong focus on democracy support in Swedish
aid over the past years, EBA finds it motivated to take a closer look at
this kind of aid based on current knowledge.
Democracy support includes a wide range of support mechanisms
aiming at developing and strengthening democratic institutions,
processes and actors. It is a broad field with many different kinds of
interventions. In this report, focus is on one specific sub-category,
support to political parties and party systems. Sweden has provided
support for these purposes through party affiliated organizations
(PAO) since 1995. The provision is guided by a special government
strategy for the period 2011-2015. Currently support amounts to
some 80 million SEK annually.
In 2015, the future of Swedish international assistance to political
parties will be on the government agenda; a new strategy is to be
drafted and decided upon. In light of this, EBA invited Lars Svåsand,
professor in Comparative Politics at the University of Bergen, to carry
out a study on international party assistance. The point of departure
for the study is to answer the question, what do we know about the
effects of international assistance to political parties in new
democracies, and to reflect upon implications for Swedish aid in this
field.
The author has made a review of the literature on research in this
field as well as of evaluations of donor interventions and programs. He
concludes that there are no clear results. There seem to be limited
effects of international party assistance, and rarely any transformative
impact, even though there are examples of positive outcomes.
However, as the report highlights there is a number of methodological
difficulties in assessing the effects. The author recognizes the
contextual challenges of carrying out support to parties and party
systems in new democracies, and underlines the need of being realistic
about what to expect from party assistance, given the relatively scarce
resources that are allocated. In addition, he points at the need for a
reasonable time perspective when assessing the results of specific
interventions.
1
International party assistance is often considered controversial, not
only because of the difficulties to show positive results but also from
the point of view that the development of parties and party systems is
a highly political and sensitive issue connected to sovereignty. At the
same time, given the key functions that political parties have in a
democracy, a number of donors still find it motivated to engage in the
development of well-functioning parties and/or party systems.
In the concluding reflections, the author provides arguments for
continued international party assistance - primarily interventions with
a cross-party orientation. As to the key factors that donors need to
consider, he underlines that ownership and commitment seems to be
necessary for success. In addition, contextual factors are very
important, and support must be “tailored-made” in order to
contribute to the higher goal of democratic consolidation.
We hope that this study will contribute to the coming discussions
and decisions on the focus, scale and mode of Swedish support in this
field. The study has been conducted in dialogue with a reference group
led by Ms Eva Lindström, former vice chair of the EBA. The
responsibility for the content of the report rests fully with the author.
Stockholm, February 2015
Lars Heikensten
Chair
2
Sammanfattning
Internationellt stöd till politiska partier (IPA) 1 kan definieras som
”organiserade insatser för att stödja utvecklingen av demokratiska
politiska partier, en god samverkan mellan partier, samt de politiska
och legala förutsättningar som krävs för demokratiska politiska
partier” (fritt översatt från definition av Burnell & Gerritts,
2010:1068).
IPA har blivit en del av de internationella insatserna för att stärka
nya demokratier och utgör därmed en del av det internationella
demokratibiståndet. Välfungerande partisystem och politiska partier
anses ha stor betydelse för ett lands demokratiska utveckling. Poliska
partier har grundläggande funktioner i ett demokratiskt system och
utgör en betydelsefull kanal för organisering och förmedling av
medborgarnas åsikter. Trots utmaningar och brister anses politiska
partier ha flera fördelar jämfört med andra sätt att organisera politiska
ståndpunkter och ansvarsutkrävande. Det finns t.ex. ingen annan
organisationsform som både kan kanalisera intressen och anta rollen
som regeringsmakt respektive opposition. Det civila samhället är
också en viktig kanal för organisering och åsiktsmobilisering men de
kan inte fullt ut anta de uppgifter som politiska partier har i ett
demokratiskt system. Givet betydelsen av ett stabilt och välfungerande
partisystem, och det faktum att man i flera nya demokratier arbetar
med att forma ett demokratiskt system och att utveckla dessa
institutioner, menar många att det finns motiv för internationellt stöd.
Samtidigt uppfattas internationellt stöd till politiska partier ofta
som kontroversiellt. Det handlar bl.a. om huruvida det i grunden är
rätt att internationella biståndsgivare engagerar sig i nationella
processer som påverkar hur politiska partier utvecklas och samspelar
med andra partier i ett visst sammanhang. Hur politiska partier
organiseras ses ofta som känsliga och högst nationella frågor. En
annan fråga i sammanhanget handlar om huruvida stödet ger någon
effekt. Det förefaller svårt att påvisa positiva effekter, även om det
finns exempel där stödinsatser har gett resultat i enlighet med
uppställda målsättningar. I sammanhanget bör dock noteras att det
finns en rad metodmässiga utmaningar med att påvisa ett samband
mellan insatser som syftar till att stödja partier och partisystem och
1
Forkortningen IPA kommer från engelskans International Party Assistance.
3
förändringar i dessa institutioner. Det handlar bl.a. om problem med
tillgång på information och data, såväl före som efter införandet av
stödet. Därtill påverkas utvecklingen av politiska partier i hög grad av
andra institutionella förändringar, såsom lagstiftningen avseende
valprocesser och politiska partier, vilket i sin tur påverkar huruvida det
internationella stödet bidrar till någon effekt eller inte.
Internationellt stöd till politiska partier utgör ett verktyg i arbetet
med att främja demokratisk konsolidering. Den övergripande
målsättningen omsätts på många olika sätt. Det finns ingen
standardmodell för hur IPA organiseras och genomförs. I denna
studie görs ett försök att beskriva de variationer som finns inom
ramen för internationellt stöd till politiska partier i form av en
diskussion kring sex olika dimensionser. Dimensionerna handlar om
syfte och inriktning av stödet, organisationsmodell för att kanalisera
stödet, den finansiella omfattningen av stödet och hur det fördelas,
vilka stödformer som används, tidpunkt för stödinsatser samt
geografiskt fokus. Exempelvis inriktas vissa insatser på enskilda partier
i syfte att bidra till partiers institutionalisering, samtidigt som andra
stödinsatser syftar till att stärka partisystemet i stort. Andra
målsättningar handlar om att stödja särskilda grupper av aktörer som
har marginaliserats i de politiska processerna. Det kan t.ex. handla om
ungdomar och kvinnor. Det är dock inte alltid tydligt hur sådana
insatser bidrar till att befästa ett demokratiskt system som helhet.
Sammantaget, visar redogörelsen av olika dimensioner att det finns
stora variationer inom ramen för de insatser som klassas som stöd till
politiska partier, något som i sig har en betydelse för hur väl
internationellt stöd bidrar till det övergripande målet om demokratisk
konsolidering.
Det internationella stödet har förändrats över tid. Det finns en
trend från att det är en stödform med huvudsakligt fokus på enskilda
partier/systerpartier till mer av partiöverskridande stöd och ansatser.
Det finns också en trend från direkt partistöd till mer indirekt stöd,
dvs. insatser som inriktas mot institutioner och processer som
påverkar partiers möjligheter att fungera på ett bra sätt på såväl kort
som lång sikt. Det handlar exempelvis om stöd till funktioner som
hänger samman med parlament, och processer som handlar om
planering och genomförande av val. Denna förändring indikerar att
IPA bör kunna kopplas närmare till andra former av bistånd.
Både forskningslitteraturen och utvärderingar av olika givares stöd
till politiska partier visar på ett antal faktorer som är av avgörande
4
betydelse för framgång. Det handlar bl.a. om betydelsen av ägarskap
och ett tydligt åtagande hos mottagarna. Därtill framhålls vikten av att
den institutionella och politiska miljön är gynnsam.
Vilka implikationer har då de iakttagelser och slutsatser som finns i
denna översikt för det framtida stödet till politiska partier? Av
litteraturöversikten är det tydligt att stödets effekter bedöms vara
begränsade. Frågan är om det är skäl för att inte ge något stöd eller om
bristen på resultat delvis handlar om de metodmässiga svårigheterna
med att mäta och identifiera resultat. En närliggande fråga är om
förväntningarna om vad som kan åstadkommas med relativt
begränsade resurser är realistiska. Samtidigt som det finns
begränsningar vad gäller stödets effekter är det av grundläggande vikt
att få till stånd fungerande partier och partisystem vilket är ett starkt
motiv för fortsatt stöd. Det finns även motiv för att mindre länder
engagerar sig i denna typ av bistånd då de kan bidra med erfarenheter
som kompletterar större givarländer på området.
I valet av inriktning finns det mycket som talar för ett större fokus
på partisystem och samverkan mellan partier. Vidare finns det ett antal
faktorer som särskilt bör beaktas för att öka möjligheterna att lyckas
med biståndsinsatserna. Som med många andra typer av bistånd,
handlar det om betydelsen av ett starkt ägarskap och institutionellt
engagemang hos mottagarna. Andra aspekter att beakta handlar om
tidpunkten för insatserna, insatsernas omfattning och huruvida det
finns processer för uppföljning, utvärdering och möjlighet till
anpassning utifrån erfarenheter under resans gång. En mycket viktig
faktor inför alla typer av insatser är att orientera sig om huruvida den
politiska miljön och de socio-ekonomiska och kulturella
förutsättningarna är gynnsamma eller inte. Avslutningsvis är det också
viktigt att se möjligheterna med att samordna med andra
demokratiinsatser och att låta stödet till politiska partier ingå i bredare
demokratistöd.
5
Summary
International party assistance (IPA) is defined as: “The organizational
effort to support democratic political parties, to promote a peaceful
interaction between parties, and to strengthen the democratic political
and legal environment for political parties” (Burnell & Gerritts,
2010:1068)
IPA has become part of the international efforts to assist new
democracies, andhence part of international democracy assistance.
Well-functioning parties and party-systems are considered to perform
essential functions in democratic systems. Political parties are
important means for organizing and channelling the views of citizens.
Despite many challenges and limitations, parties have advantages over
other ways of organizing accountability in democracies. No other
organization performs the critical functions of interest aggregation
and the organization of government and opposition alternatives.
While civil society organizations also are necessary in democracies,
they perform different functions from those of parties. Given the
importance of a stable and well-functioning party system, and the fact
that many new democracies struggle with developing these
institutions, many find it motivated to provide international
assistance.
IPA is nevertheless controversial for several reasons. It is a
normative issue if it is correct for donor institutions to be involved in
processes that shape the nature of political parties. The types of and
the organization of political parties are among the most ‘national’
sensitive issues in politics. IPA has also been controversial because it
has proved hard to demonstrate that it has positive effects. Many
studies of IPA concluded that there are limited effects. But there are
also examples where IPA has functioned according to specified
objectives.
There is a wide range of methodological challenges in detecting the
relationship between efforts to assist parties and party systems, and
changes in these institutions. The issues here involve problems of
access to data on several time points; prior to and after the
introduction of IPA. Party developments are also strongly embedded
in a wider setting of institutional environments, such as electoral
processes and legal regulations that impact on political parties;
6
institutions which also are recipients of democracy assistance. IPA’s
effects are likely to be influenced by these environments.
IPA is an instrument in assisting new democracies to consolidate.
This overall objective is operationalized in diverse ways. There is no
standard model for how IPA is organized and implemented. The paper
discusses six dimensions of IPA: the objectives of IPA, the
organizational model, the financial scale of intervention, the modes of
intervention, the timing of intervention, and the geographic focus for
IPA. Each of these represents groups of variables that may influence
how IPA can contribute to the overarching goal of democratic
consolidation. Some IPA projects targets individual parties, where the
objective is party institutionalization, but at the same time another
objective is to assist the development of a stable party system, which
has different properties from those of individual parties. An
additional objective in IPA has been to support particular groups of
actors, such as youth and women, who have been marginalized in the
political processes. The different ways of organizing IPA reflect these
dualities, and it is not always clear how each contribute to the
consolidation of democracy as a whole.
Across time there is a trend away from party-to-party linkages
towards more cross-party approaches. There is also trend away from a
concentration on direct party support to indirect party support; that is
IPA programs reach beyond parties to institutions and processes that
impact parties in the short or long run, such as support for
parliamentary institutions, electoral management bodies and civil
society organizations. This change indicates that IPA programs should
be closer linked to other forms of development assistance. Both the
research literature and the evaluations of IPA programs indicate that
critical factors that influence the ‘success rate’ of IPA are, first, the
‘ownership’ issue, such as when IPA programs establish trust among
the participants and where the recipient parties are committed to
change, and second, the institutional and political environment in
which IPA is situated.
What are the implications of the findings and conclusions in this
reviw for future assistance to political parties? From the literature
review, it is clear that the effects of IPA are considered to be limited.
Does this imply that assistance should not be provided, or is the lack
of effects mainly a matter of methodological difficulties in measuring
and identifying results? A related issue is whether the expectations on
what should be achieved are realistic given the relatively limited
7
resources. Even though, there are limitations in terms of effects of
IPA, it is still essential to develop stable parties and party systems,
which is a strong motive for continued support. There are also
arguments for why smaller states should be involved in this field as a
complement to some of the major donors, When choosing direction it
seems motivated for donors to focus more on party-systems and
interparty dialogue. If IPA is continued there are several factors that
should be observed to increase the likelihood that IPA programs
succeed. These are partly characteristics of IPA programs themselves
such as: ownership, institutional commitment, the time dimension of
projects, the financial scale projects, ensuring monitoring, evaluation
and adjustments. However, IPA programs must also take into account
contextual factors such as unfavorable political environments and
unfavorable socio-economic and cultural conditions. IPA should also
be part of a broader democracy assistance agenda which calls for
coordination with other programs.
8
1. Introduction
An increasing number of countries have introduced programs in
support of political parties in new democracies. International party
assistance (IPA hereafter) has been defined as: “The organizational
effort to support democratic political parties, to promote a peaceful
interaction between parties, and to strengthen the democratic political and
legal environment for political parties”(Burnell & Gerritts, 2010:1068).
IPA is, just as development aid in general, nevertheless controversial,
mainly because of the normative question if this is something external
actors should become involved in at all, but also because one does not
know if IPA has the intended effects or not, or if there are significant
unintended effects.
The purpose of this study is to review the findings of research on
this topic and evaluation reports of donor institutions. Hence, the aim
of the paper is to provide an answer to the question: What do we know
about the effects of international assistance to political parties in new
democracies? And, as a follow-up to that question: which steps should
IPA take if it is to be pursued in the future? The study will also
include description of different dimensions of IPA, and how some
different donors channel their support.
There are other, relevant questions to pose regarding IPA which I
deal only briefly with. First, there is the normative issue: should states
and multilateral institutions that provide development aid be involved
in this field at all? This is an important debate, worthy of a study in
itself. I briefly discuss this issue, as the question is relevant for the way
in which IPA has been organized, and return to the normative issue in
the conclusion. A second issue concerns the effects of IPA on parties
in the donor countries that are involved. IPA can have various
consequences for donor parties. A positive impact of IPA is that it
may increase awareness and understanding of democratization
processes. Party activists participating in IPA projects may therefore
be better informed about development issues in general. But it is also
possible that IPA could contribute to centralization inside the donor
parties, unless IPA is properly anchored and the projects shared in the
donor party organization. Although the impact on the donor parties is
interesting and worth to study, a discussion of this topic would
require data that, as far as I know, are not available.
9
1.1 Democracy assistance and party assistance
Democracy assistance has grown considerably since the early 1990s
(Cornell, 2013) and reflects the understanding of democracy as a
multidimensional concept and composed of different types of
institutions. Although there is no single definition of what a
democracy is, most analysts would agree that a democracy includes
some elements that are necessary, but not sufficient for a system to be
called a ‘democracy’ 2. Democracies include some form of vertical and
horizontal delegation of authority. Vertical delegation is the selection
of a group of representatives, among a set of candidates competing on
a level playing field, while horizontal accountability involves a division
of power among several institutions (Bergman, Muller, & Strøm,
2000; Schedler, 1999). A democracy also has judicial framework
guaranteeing the rights of individuals and groups.
Significant efforts in democracy assistance have been directed to
improve the electoral process (Darnolf, 2011; Kelly, 2012; Kuncic,
2011; Lean, 2007). Other programs have sought to strengthen the
judiciary (Risse, Magen, & McFaul, 2009), the parliament (Burnell,
2009; Hudson & Wren, 2007; Power, 2008), the government
administration (Fritz & Menocal, 2007), civil society (Hearn, 1999),
mass media (Kumar, 2006) or promoted decentralization (Romeo,
2003).
With some variation, the main bilateral and multilateral donors
assisting the development of democracy have concentrated their
attention on three principle areas: improving electoral processes,
strengthening institutions, such as parliaments and the judiciary, and
civil society. Support for political parties as such was added to the
democracy assistance agenda late, primarily because support for
political parties was seen as interference in the domestic politics of the
recipient states.
Democracy assistance focused initially on strengthening civil
society, also considered to be vital for democratic consolidation.
However, strengthening civil society needed to be balanced by
strengthening political parties (Doherty, 2001). On the one hand,
many civil society organizations (CSO hereafter) did not see
themselves as political actors, therefore political parties needed to fill a
2
See (Dahl, 1971)
10
void. But on the other hand, a strong civil society could potentially
increase the ‘demand’ side in the political system. Improvement of
political parties would be necessary to take care of interest aggregation
and policy coordination. Only political parties perform these two
functions in political systems.
1.2 Method, material and limitations
The findings and conclusions of this paper are based on a desk review
of the current knowledgebase for international party assistance.
A main source of information for this paper is the evaluation
reports that the IPA institutions have commissioned from
independent experts as well as the growing research literature on party
assistance. There are an increasing number of actors involved in IPA
(see list of acronyms). Some institutions have been active since the
1960’s while others were formed during the last years. The availability
of information and evaluations varies enormously, both between
institutions as well as over time for the individual institution. Some
IPA donors, like NIMD (The Netherlands Institute for Multiparty
Democracy) has a long series of evaluation reports. Other IPA
donors, like the British Westminster Foundation for Democracy
(WFD) and the two American institutions, National Democratic
Institute (NDI) and International Republican Institute (IRI) have
some evaluation reports, while no evaluation reports are available for
the German foundations even if they are among the largest and oldest
IPA institutions. It is, in some respects, a skewed sample where those
institutions that have made available reports in English are clearly
overrepresented.
Geographically, there is an overrepresentation of reports on Africa
and the former Communist governed countries, but these regions are
also among the top beneficiaries of IPA. The geographical skewedness
may impact on what general conclusions that may be drawn, but this is
in any case problematic because of the significant impact of the context
for IPA projects. The context refers both to the timing of IPA
projects, that is when IPA projects are introduced (prior to, or during,
or after the transition to democracy) as well as the particular
configuration of environmental factors, such as the structure of
political institutions in the recipient countries, and more general
socio-economic conditions. To some extent every context is unique.
11
Nevertheless, the types of IPA donor institutions and their activities
that are discussed also identify several issues that re-appear across the
reports which indicate that there are some factors that are important
when drawing conclusions about what works or not.
IPA has not been a subject of great concern among political
scientists, but there is a growing interest in it. Peter Burnell (Burnell,
2006; Burnell & Gerrits, 2010) and Tom Carothers (Carothers, 2005,
2006) have spearheaded the research in this field and to some extent
the research literature has compensated for the lack of evaluation
reports by the German IPA actors (Erdmann, 2006; Weissenbach,
2010, 2013, 2014). Research on IPA covers new democracies in all
parts of the world, but a majority deals with the former Communist
bloc and Sub-Saharan Africa.
1.3 Outline
I first outline why IPA has become part of the democracy assistance
agenda and tie this discussion to the perception of parties as necessary
institutions in democracies. This is followed by a discussion of some
of the problems confronting IPA, particularly that the types of parties
emerging in many new democracies are fundamentally different from
those in established democracies. After that I discuss six dimensions
of IPA programs; the objectives of IPA, the organizational model, the
financial scale of intervention, the modes of intervention, the timing
of intervention, and the geographic focus for IPA, There is no
standard model for how IPA is structured. Different donors have
focused and organized their assistance in different ways.
In the following section I discuss the key question of this study,
i.e. findings on the effects of IPA, A main problem in the study of
IPA is to identify if IPA has the effects it is supposed to have. I
therefore first discuss the methodological difficulties in analyzing
IPA’s contribution to democratic consolidation and after that I
present and comment upon what we know from the literature on
parties and evaluation reports of IPA programs.
The final part of the paper returns to the normative issue and
arguments are presented for why IPA should be continued; in spite of
the problems in detecting clear effects. I summarize what I see as
critical factors that should be taken into account in the design of IPA
programs.
12
2. International Party Assistance –
motivation and challenges
2.1 The necessity of parties in democracies and the
background for IPA
The dominant view in political science is undoubtedly that political
parties are ‘indispensible’ institutions in a democracy (Russel J. Dalton
& Wattenberg, 2002; Lipset, 2000; S.C. Stokes, 1999). This is an
opinion held not only by political scientists, but also by citizens in
general. Although citizens in established democracies have become
increasingly critical of their parties, they nevertheless - to an
overwhelming degree - support the principle that political parties are
necessary in a democracy (Russel J. Dalton & Weldon, 2005).
The ‘necessity’ of political parties is also reflected in various public
policies. The introduction of public subsidies for parties in established
democracies has been justified, among other reasons, because these
institutions were seen as essential for the functioning of the
democratic system. Aid to political parties has also been integrated
into international programs aimed at strengthening democratic
governing institutions and processes in newly democratised states
because, as one of the donor institutions puts it, political parties are
“…a cornerstone of representative democracy and serve a function like no
other institution” (NDI, 2008b).
There are several reasons for the claim of a link between parties and
democracy.
First, there is an empirical justification in the sense that there exists
no fully fledged democracy today that does not have political parties.
Since the ‘third wave of democratization’ there are more political
systems that can be called ‘democratic’ today than at any other time
and the number of political parties has never been higher. But parties
in newer democracies are characterized by instability, leadership
fixation, and weak capacity to perform the functions associated with
parties in democracy.
Second, although there is a range of additional ways of organizing
the political process in democracies, such as referenda, citizeninitiatives, consultation forums, deliberative polls, recalls etc, (Russell
13
J. Dalton, Cain, & Scarrow, 2006; Smith, 2009), all of these are
supplementary alternatives to the representative democracy in which
political parties have a dominant role, not substitutes for them. There
does not seem to be any alternative that can substitute what parties do,
particularly regarding coordination of decision-making in parliaments
and the need for ‘interest aggregation’ (Muller, 2000).
Third, there is also an argument for political parties based on
democratic theory. Regardless of whether parties are necessary or not,
they nevertheless will exist in a democracy because of the right of
individuals in a democracy to freely form and join organisations;
including political parties. Because citizens can freely form and join
political parties, the political party has emerged as superior to other
ways of solving collective-action problems (Aldrich, 1995). In all
political systems it is necessary to delegate power from the citizens to
a small group of representatives, also known as ‘vertical accountability’
(Przeworski, Stokes, & Manin, 1999; Schedler, 1999). Political parties
are seen to provide better ways of ensuring vertical accountability than
any other alternative. This is because it is easier for voters to be
informed about the views of candidates running as candidates for
political parties that appear in several elections across time, than to be
informed about the political views of a range of individual, nonpartisan candidates (Mitchell, 2000).
Political parties also have an advantage in creating lasting alliances
of representatives that form the basis for government and opposition,
in contrast to the ad-hoc creation of groups which would be necessary
in a parliament consisting only of non-partisans. The political party, as
an organization, is also likely to endure over time, independent of the
specific individuals in the organization. Thus, by appearing in several
elections, voters are better equipped to evaluate to what extent
representatives of political parties have delivered what they promised
or are likely to continue to do so or not. Political scientists further
explain the prevalence of parties in democracies by the type of
functions they perform, particularly interest aggregation and forming
the basis for government, alternatively opposition. It is the
performance of these functions which make them necessary
institutions in a democracy.
The ‘inevitability’ of a link between political parties and democracy
has been derived from the analysis of how the political party gradually
became an important collective actor in the development of Western
democracies. The significance of political parties in established
14
democracies was linked to the ‘passing of thresholds’(Rokkan, 1970),
such as the right to opposition, freedom of the press, the widening of
the suffrage, extension of representation through proportional
electoral systems and finally the introduction of parliamentary
government. Most of the literature sees political party development in
tandem with democratization in general, particularly the extension of
the suffrage (Duverger, 1967), while the structure of the party system
was impacted by the politicization of cleavages (Lipset & Rokkan,
1967) and by institutional arrangements, particularly the choice of
electoral system (Colomer, 2005). True, there were also break-downs
in the process (Germany, Italy, Austria, Spain), but most countries
developed gradually towards some form of consolidated democracy.
Thus, political parties were partly the cause of improvements in
democracies, as in the fight for suffrage extension, partly the
consequences of the extension of democratic rights. A dominant view
explaining the emergence of stable party systems in what is now
established democracies is that the politicisation of cleavages in
society (along territorial, economic and cultural cleavage lines)
generated political parties reflecting such cleavages and thereby
incorporated large segments of the population into the political
process (Lipset & Rokkan, 1967).
Nevertheless, in spite of the strong and widespread claim of the
importance of parties in democracies, it is primarily still a claim, but
hardly demonstrated (Biezen, 2003). Parties alone cannot prevent
democratic regress or collapse. Parties function alongside a network of
political institutions and in a social, cultural and economic context.
Some scholars, such as Schmitter for instance, is rather pessimistic
concerning the state of parties in new democracies: “one of the major
reasons that I am so convinced of the basic weakness of parties in
these neodemocracies is that virtually all the difficulties that they have
been experiencing are also being experienced by contemporary parties in
achaeodemocracies” 3 (Schmitter, 2001:84).
In contrast to the (mainly) gradual democratization in Western
Europe, the transition towards democratic governance in the “Third
wave” 4 was often compressed into a very short time period. In many
3
Ithalics in the original.
«The Third wave” refers to Huntington’s distinction between major periods of
democratization: First wave,1828-1926, Second wave 1943-62, Third wave: 1974-, starting
with the transition in Portugal (Huntington, 1991)
4
15
countries the transition resulted in uneven progress and subsequent
variations in the extent of democracy, as exemplified in the various
scales of democracy, such as Freedom House scores, and in various
categories, such ‘liberal democracies’ and ‘electoral democracies’.
Analysts of non-European areas point out that party developments in
other regions are likely to be different: “The European experience is
not a yardstick of ‘normal’ political development” (Rodan, 2012: 313).
It would be surprising if the status and nature of political parties in
new democracies were not affected by the way democratization
happened, by the speed of democratization, and by the radically
different cultural, economic and other contextual factors. The linkage
between democratization and other processes of change in many new
democracies are fundamentally different, coined by Rose and Shin as
“democratization backwards” (Rose & Shin, 2001). Elites and masses
in new democracies have not had the same opportunity as their
counterparts in Europe to gradually adapt to changing political
institutions and social and economic processes. Instead, political
transitions have been compressed in time and occurred simultaneously
with economic transformations, technological changes and a strong
impact of international events, actors and processes.
In sharp contrast to the democratization in what is now established
democracies international factors have come to be seen as important
explanatory factors for the emergence of the new democratic regimes
in general (Brinks & Coppedge, 2006; Levitsky & Way, 2005; Levitsky
& Way, 2010; Resnieck & Walle, 2013), and for the development of
political parties, particularly in Eastern and Central Europe (Lavenex
& Schimmelfennig, 2011b; Pridham, 1990; Stewart, 2009). It is against
this back-drop that IPA has been introduced.
2.2 The challenges of parties and party systems in new
democracies
There are usually two objectives for party assistance programs: to
strengthen political parties and to assist in the development of a
functioning party system. In both cases, there are a number of
challenges in new democracies. A political party is a complex
institution and the party system is more than the sum of the individual
parties. Both individual parties and the party system interact with
other institutions and processes.
16
Political Parties
Political parties, as they are structured in established democracies, are
institutions with multiple levels and multiple organizational units. The
party organization is a hierarchy with units at the local and regional
levels connected to the national level. Parties vary in how extensive
they have been able to penetrate the territory with organizational
units. In addition to the party organization proper, there are often
auxiliary organizations for students, women, youth, and senior
citizens. At each level there are also groups of party representatives in
local, regional and national representative institutions. The
relationships between these units, horizontally as well as vertically, do
not follow a standard format, but is a mixture of party history and
constitutional features.
Political parties in democracies depend on a level playing field, an
impartial administration of the electoral process as well as on the
availability of multiple mass media channels. A judiciary able to
uphold the constitution and the legal acts affecting political parties,
civil society, elections and the media has a positive impact on political
actors. As political parties are the prime actors in elections, the balance
between political institutions with party representatives, such as local
and regional bodies, parliament, government and president, also
impact on the internal dynamics of the political parties.
An important component of an institutionalized party is that the
party operates according to a set of statutes which structures how the
party carries out the standard ‘functions’ of parties, such as
nomination of candidates, leadership selection and policy formulation.
Implicit in the concept of party institutionalization is that the
participants in the party share common political perspectives, an
ideological orientation for short, which is why the participants are in a
particular party and not with just any party. It is assumed that in an
institutionalized party the participants share identification with the
party and that therefore party elites, in particular, tend to be loyal
participants in the party over a long period, advancing from regular
grass-root members to leadership positions. Yet, what specific
characteristics institutionalized parties have are not always clear, even
in established democracies. Take the contrast between the Democratic
and the Republican parties in the United States vs. the Social
democrats in Sweden and in Norway. The two US parties are
obviously institutionalized in the sense of having long-lasting
17
organizations, as are the two major parties in the two Scandinavian
countries. Yet, in terms of organizational characteristics there are as
many dis-similarities as similarities between the US parties and the
Swedish and Norwegian examples 5.
Thus, while there are significant variations among parties in
established democracies, the notion of ‘party’ is more or less
understood. This conception of party is not prevalent in many new
democracies (Svåsand, 2013) and this has been raised as a criticism
against IPA as a whole. IPA has been criticized for promoting an
‘ideal’ model of a party that does not exist anywhere (Carothers,
2006), least of all in many new democracies. IPA is based on an
(implicit) assumption that political actors are united in parties that
advocate collective interests, such as for social classes, religious,
linguistic or territorial groups, or mobilize support for the promotion
of particular values or issues. For this type of parties, actors are
assumed to identify with their parties’ ideas and that they have an
interest in pursuing party building in the long run.
However, a frequent critique of parties in new democracies is that
many of them are created by political entrepreneurs for their own and
their families’ benefit and dominated by a small group of leaders who
are not interested in party building 6. That a certain type of
organization has been chosen because it serves the interests of
ambitious politicians has figured prominently in the study of Russia.
Stoner-Weiss, for instance, argues that “Under the conditions of dual,
simultaneous economic and political transitions, elites do not
necessarily flock to the safety and predictability of political parties.
Rather, they may prefer an equilibrium of political underinstitutionalization to preserve their early winnings from the
economic transition” (Stoner-Weiss, 2001). Hale argues similarly:
“creating or building parties is a choice if they are considered better
instruments than other substitutes for winning office(Hale, 2007), and
to McFaul ”..parties are weak in Russia because the most powerful
politicians have made choices to make them weak”(McFaul,
2001:1160). Parties dependent on individuals and/or their close
5
The title of Richard Katz and Robin Kolodny’s (R. S. Katz & Kolodny, 1994)chapter on
American parties in a comparative volume on party organizations, is indicative: “Party
organizations as an empty vessel: Parties in American politics” .
6
It would be incorrect that all parties in new democracies are of this kind. Some have grown
out of liberation movements. However, see (Hyden, 2006) for arguments for why such
movement based parties have not made the transition to democratic organizations.
18
relatives also flourish in other regions, such as in Korea: “:… the
political parties of the three Kims were nothing but screens to hide the
informal patron-client links between politicians and voters”, according
to Hellman (Hellmann, 2011:476), while in the Philippines Aceron
calls parties temporary political alliances, or ‘fan clubs of politicians’
(Aceron, 2009:5). Some analysts of Nepal argue that all political forces
“continued to be overshadowed by the individual personalities of the
upper-class coterie dominating party politics. In terms of political
favor, the expectation (and reality) is that politicians will give priority
to vested interests, looking after their immediate family, then their
close friends, then the local community, and lastly the people in
general” (Thapa & Sharma, 2009:216).
Thus, while democracy may have become ‘the only game in town’
(Linz & Stepan, 1996), it does not imply that even if election is the
only means to gain control of the government, it will be preceded by –
or followed by - extensive party building.
Many parties in new democracies are characterized as being topheavy with weak internal democracy. One of the objectives of IPA has
been to improve – and to change - party organizations, including
internal party democracy. The reports from individual donor
organizations all identify this issue as one of the most difficult to
work on: ”…advancing internal party democracy is frought with
potential pitfalls” (IRI, 2010: 25) This is not only because it is difficult
to document change in internal democracy 7, but primarily because
increased internal party democracy is seen as a challenge to the
position of established elites. These obstacles are another reason why
IPA donors have expanded their targets towards civil society
organization and young political activists in order to encourage the
development of an alternative political leadership in the future.
Parties for this type of leaders are vehicles for advancements of
personal interests and can therefore be abandoned if the actors believe
that jumping to other parties will be more advantageous 8. For IPA to
succeed in building strong party organizations one needs to understand
what the incentives are likely to be for actors to have a long term view.
Both IPA and democracy assistance in general – and evaluations of
7
See (Berge, Poguntke, Obert, & Tipei, 2013) for how this could be identified.
Defection among parliamentarians from the party they were elected to represent to other
parties is a big problem in many new democracies. See for example (Ames & Power,
2010:204-205) for Brazil and (Young, 2014) for Malawi.
8
19
such programs - need to have a realistic view of what it is possible to
achieve, given the context for the assistance and the available means.
Party Systems
IPA can also seek to improve the party system. The qualities of the
party system are such as: the interaction pattern between parties, the
number of relevant parties, the ideological polarization of the parties
and the relative size of the parties. The reason why IPA is targeting
the party system is because it is not the individual parties that are
crucial for democratic consolidation but the nature of the party
system. It is widely recognized that the development of a functioning
party system is an ‘Achilles heel’ in new democracies.
Nevertheless, it is not obvious which qualities of the party system
are most conducive for democracy. Although multiple parties are
necessary for a democracy, it does not follow that the more parties
there are, the better. At some stage, too many, almost equally sized
parties, combined with ideological polarisation 9, are deemed to be
negative for democratic governance. When the number of relevant
parties, those needed either for coalition building, or which has
blackmail potential, exceeds five, which Sartori classifies as a highly
fragmented party system and when ideological division is one of
polarization, the prospect for democratic survival diminishes (Sartori,
1976). His comment on Chile in the run-up to the 1973 election is
worth quoting: “On the sole ground of the acceleration of its centrefleeing polarisation, it was an easy prediction that all the conditions of
democratic governance were rapidly dwindling. And the fact that this
very obvious point escaped not only the actors but also the observers
represent an ominous symptom” 10 (p.144). This argument is shared by
Peeler (1998) who in his summary of ten elements impacting on
democratization in Latin America states: “Party systems that are
institutionalized and characterized by moderate pluralism are most
conducive to liberal democratic stability…. Conversely, a party system
with a proliferation of short-lived parties representing only ambitious
9
As Dalton has shown, ‘fragmentation’ and ‘polarization’ are two separate characteristics of
party systems that do not necessarily correlate (Russel J. Dalton, 2008)
10
However, he also identifies other variables for countries that experienced democratic
breakdown (Weimar, the French 4th Republic, Spain and Italy in the interwar period) such
as constitutional and economic variables.
20
elites and with few popular roots, is a recipe for un-governability”
(Peeler, 1998: 192).
Thus, highly fragmented party systems may lead to difficulties in
establishing governments able to decide and to implement policies.
The result could be political instability and ultimately to problems in
sustaining democracy under problematic conditions. Similarly,
polarized party systems lead to lack of continuity in policies and
creating an adversarial political atmosphere which may lead to
democratic collapse. However, it is also argued that the opposite of a
polarized party system, one in which there is no significant differences
between parties, is equally harmful to democracies. In such systems,
the electorate is not able to understand what policy alternatives the
parties offer. Lack of policy differences may lead to politicians to
defect from one party to another in search for personal advantages and
hence, breed distrust among voters and elites. Lack of programmatic
differences between parties is associated with other types of linkages
with voters, such as clientelism and corruption (IIDEA, 2014b;
Keefer, 2005)
Another quality of the party system, which might be a hazard to
democracy, is the development of a dominant party system. This is the
case if one party wins a supermajority of the parliamentary seats
and/or the presidency in at least three consecutive elections
(Bogaards, 2004). Although there are several examples of democracies
that have been governed by dominant parties (Sweden, Japan and
Italy) it has been argued that in developing countries the new
democracies lack some of the qualities of established democracies,
such as a constrained executive power, effective expression of
minority interests and a secure electoral process where the outcome of
the electoral process is uncertain (Giliomee & Simkins, 1999). Thus,
in a new democracy, such as in South Africa, the rise of a dominant
party system is seen as ominous sign for democratic consolidation
(Southall, 2001).
A possible countervailing factor could be
decentralization which will allow opposition parties at the national
level to control parts of the political system and be able to develop a
competitive capacity.
Thus, as regards the party system and democratic stability, or
consolidation, there are at least three characteristics that should be
avoided:
21
•
a high level of fragmentation; meaning more than five relevant
parties represented in parliament,
•
a high level of ideological polarization, particularly between
the largest party alternatives,
•
a dominant party system in a political system combined with
weak institutional restraints on the executive power.
The structure of the electoral system, the management of elections
and the nature of electoral campaign impact on the representation of
parties. The parties, through policy orientation, can choose to engage
in compromises and cooperation patterns with each other. Hence,
changes in the party system are partly a consequence of actions taken
by the parties, and partly an effect of the environment and changes in
the environment.
22
3. Dimensions of International Party
Assistance
There is no ‘standard format’ for how IPA has been organized and
implemented. In the following section I outline what I see as the main
dimensions of IPA programs. Within each dimension I highlight the
characteristics of IPA donors and in the end discuss the extent to
which their objectives have been reached. The information under each
dimension is extracted from evaluation reports of donor institutions.
In appendix B, I provide an extensive summary of such reports which
contains also specific references to the findings.
IPA varies along several dimensions:
•
objectives of IPA,
•
organizational model,
•
financial model and scale of intervention
•
modes of intervention
•
timing of intervention,
•
geographic focus,
The first three concern the IPA programs as such, the fourth one –
modes of intervention is partly linked to the two first, but also
impacted by the latter two which relate to the context for IPA
interventions. I believe it is because IPA varies along all these
dimensions that it is difficult to demonstrate that one ‘model’ is
superior to other models or to assume that what works in one setting
will also work in other settings.
3.1 Objectives of IPA
A key issue in IPA is to identify what IPA is meant for. The
legitimation for why IPA has been introduced is the view that parties
are necessary institutions in a democracy. IPA is therefore supposed
to contribute to democratic consolidation. The decision to establish
and to organize IPA in a certain way is taken by political actors in the
donor countries in order to achieve certain goals. IPA has been
23
pursued with at least five objectives, which in various ways, it has been
argued, all contribute to democratic consolidation:
•
securing a particular political outcome,
•
strengthening single parties,
•
support for multiple parties,
•
strengthening the party system, and
•
targeting sub-groups of political actors.
Securing a political outcome
In some cases IPA has had an explicit political purpose: IPA has been
meant to result in a specific political outcome. The American NDI and
IRI institutes, for example, were actively promoting the opposition
parties in Serbia at the end of the 1990’s. The objective was to bring
the Milosovic-regime to an end. This effort has been referred to as
“….one of the great success stories of political party assistance and
democracy assistance generally” (NRC, 2007:37)11. Another example
is the long term assistance to ANC in South Africa and to SWAPO in
Namibia, by the Friedrich Ebert Stiftung (FES); the party foundation
of the Social democratic party in Germany (Vinnai, 2007: Ch. 7).
These efforts have been justified because of the higher goal, promoting
a democratic system. Implicitly at least, similar goals are part of many
other IPA programs. In Uganda for example, several donors have
coalesced in joint programs; first in the Deepening Democracy
Program (DDP) 2008-2011, and since then in a new program,
Democratic Governance Facility (DGF). While the programs are
nominally neutrally formulated, a motivation among the donors has
been to improve the state of the opposition parties; to make the party
system more competitive. In other words: to reduce the dominance of
the incumbent party, the NRM-O. 12
11
12
See also (Spoerri, 2010)
See (Power & Coleman, 2011:31ff)
24
Strengthening single parties
More common though, are IPA programs that seek to improve
specific individual political parties. This is often called ‘the sister party
model’, and is a main element in the Danish, Swedish and British IPA.
But the term ‘sister party’ model is increasingly a misnomer.
Originally, it referred to IPA programs in which partnerships were
established between parties that belong to the same ideological family.
But as IPA programs have expanded geographically from Central and
Eastern Europe, where the cleavage pattern is often similar to Western
Europe and into non-European new democracies, it is not always
possible to identify parties in the same way. Party-to-party approach is
a more appropriate label. Party-to-party cooperation is a significant
part of IPA in the three mentioned countries. In Sweden 70% of the
funding for IPA is allocated to such projects, while the corresponding
share in UK and Denmark is 50 and 40%, respectively. Other donor
institutions, such as the two American and the German foundations
are prohibited from their financial sponsors to support a particular
party 13. The essence of this approach is that partner parties are chosen
by the donor parties themselves (within certain broad constraints,
such as parties’ democratic orientation). One of the arguments in
favor of this model is that it makes a distinction between IPA
programs and the foreign policy institutions of the donor country. But
as the evaluation reports for the Swedish IPA program shows, there
are many challenges for party foundations to manage such programs,
although improvements have also been made as the foundations have
become more routinized.
There are two other issues that the party-to-party party model has
struggled with. First, it is not evident that the partners that donor
parties choose are linked to the overall aim of party system
consolidation or to democratic consolidation. 14 Second, an increasing
concern among donors is the linkage between party support and other
types of democracy assistance. Party-to-party support may become
disconnected from other types of democracy assistance, or
development assistance in general 15. In Denmark, for example, the
parties cooperating in DIPD have projects in 14 countries (Bhutan,
13
The German foundations were originally focused on sister parties. Although this is the
general rule in the US, USAid sometimes waves this requirement.
14
See the reports on Swedish IPA in appendix B.
15
A critique found in many, but not all, of the evaluations of Swedish IPA.
25
Bolivia, Egypt, Ghana, Honduras, Kenya, Malawi, Myanmar, Nepal,
Palestine, Swaziland, Tanzania, Zambia, and Zimbabwe), but there is
minimal overlap with countries receiving most of Danish ODA. Only
three countries, Bolivia, Tanzania and Zambia, are among the 14 most
prioritized countries for ODA(DIPD, 2013:15) 16. Sweden is an outlier
in this regard, compared to other major donor countries. The share
devoted in party-to-party project has been higher than elsewhere, but
over time, the multiparty share has increased.
Support for multiple parties
Strengthening individual political parties is usually also part of the
objectives for IPA. A goal for United States democracy assistance to
‘establish viable democratic parties’ (United States Agency for
International Development) (USAID 2003), while the Netherlands
Institute for Multiparty Democracy (NIMD) have projects in
“support for institutional development of political parties” (NIMD,
2012).
For multilateral institutions involved in party assistance, such as
IIDEA (International Institute for Democracy and Electoral
Assistance) and UNDP (United Nations Development Program), it is
almost by definition that their orientation is towards assistance for
multiple parties, but this is, as we saw above, also increasingly the case
for national donor organizations, including those that have also a
party-to-party component (such as the Danish Institute for Parties
and Democracy (DIPD) in Denmark, the Westminster Foundation
for Democracy in the UK, and the Swedish PAOs).
All of the major IPA institutions include several parties. The
argument for this approach is that it avoids associating the donor
institution with taking sides in the domestic political process in the
recipient country. It takes the existing configuration of parties as a
point of departure and provides opportunities and resources that in
principle are available to’ all’ parties. ’All’, however, does not
necessarily mean every party that is registered. IPA donors select
parties by means of various criteria, such as parliamentary
representation, or electoral support in local or national elections. In
16
The others being Afghanistan, Bangladesh, Burkina Faso, Ghana, Kenya, Mozambique,
Nepal, Somalia, South Sudan, Uganda and Vietnam.
26
practice, the field offices of the donor institution have some
autonomy in choosing which parties to include. The inclusion of
several parties means that IPA does not express a preference for one
or the other party; not explicitly at least.
Nevertheless, there are also problems with this approach. The
principle in NIMDs approach is that all parties with parliamentary
representation are invited to participate in so-called inter-party for a
(see below on modes on intervention). This includes obviously the
incumbent party. In dominant party systems, as in Uganda, the
inclusion of the ruling National Resistance Movement-Organization
(NRM-O), may actually help entrench the dominant party in its
position, and thus delay the emergence of a more competitive party
system. (In both Mozambique and in Georgia prior to the election of
2012 17 the party alternatives to the incumbent party have been so weak
that other actors, civil society groups and minor parties have had to be
included in the programs). When the International Republican
Institute (IRI) tried a similar approach in Azerbaijan it backfired as
the opposition parties could not accept that IRI also helped the party
they were struggling against. However, by excluding the incumbent
party an IPA donor risks alienating the regime which may brand
opposition parties as ‘foreign agents’.
Strengthening the party system
Strengthening the party system, rather than only individual parties is
usually also part of the objectives of IPA actors. Thus the initial
Swedish IPA program sought to “….contribute to the development of a
well-functioning and pluralistic party system and democratic societies in
developing countries and in Central and Eastern Europe 18” (cited in
(Uggla et al., 2000: 3-4), or as formulated by the Dutch NIMD :
“Supporting the process of democratization in young democracies by
strengthening political parties and political groups as bearers of democracy
in order to create a well-functioning, sustainable pluralist political party
system.” (NIMD, 2007) 19 The argument in favor of a focus on the
17
The 2012 election overturned completely the dominance of the United National
Movement
18
This, and other quotations from the report, is my own translation
19
NIMD also has a third objective: Improving the relationships between civil society and
political parties(NIMD, 2007: 7)
27
party system is the perception that a stable party system is linked to
the consolidation of democracy. Reducing electoral volatility and
structuring the interaction between parties are important features of
party system institutionalization. 20 Strengthening the party system
includes, but is not limited to, strengthening the individual parties.
The most explicit focus on the party system is that of NIMD where
the inter-party forum is a mechanism to facilitate inter-party dialogue,
but also other donors organize arenas for multiple parties to discuss
issues. Only NIMD, however, has established such an arena as an
institution.
Given the properties of a party system (number of parties, relative
size of parties, ideological polarization of parties, and the interaction
pattern between parties), efforts to strengthen the party system
inevitably have to look beyond the parties themselves. This includes
efforts to reform and improve the electoral management system in
order to have a level playing field for parties and candidates, reform of
the legal regulations of parties, and the structuring of public subsidy
schemes for political parties. These fields have been of particular
concern to multilateral institutions like IIDEA and UNDP.
It is recognized that the structure of the party system can have a
direct impact on how the political system functions. But to impact on
the party system, IPA must also consider elements outside of the
parties.
Targeting sub-groups in parties
Finally, IPA- institutions, both multilateral and national donors, also
target sub-groups of party organizations or groups of political
activists. In particular, there has been an emphasis on programs
seeking to increase the recruitment of young people and women in
politics, two groups severely underrepresented in political institutions.
20
Casal Bertoa (Bertoa, 2014) has shown in a study of 60 European democratic regimes
between 1848 and 2013 that stabilization of the party system is important for democratic
consolidation, while party institutionalization is not. Whether or not this relationship also
holds for political systems in other world regions is yet to be seen. But his study includes
also the new democracies in post-communist Europe that were recipients of IPA. IPA
however, is not a variable in his study. The study is nevertheless relevant for IPA as it argues
that party system institutionalization is a more important factor in democratic consolidation
than party institutionalization.
28
Many IPA donors have turned their activities away from, or least as
supplement to, projects targeting the party organizations directly. The
German Konrad Adenauer Stiftung (KAS) in Malawi for example
started a so-called Young Politicians Union (YPU) in 2004. The
purpose was to provide an organization recruiting young political
activists from different parties for work on joint projects. By targeting
this group, the initiative avoided the problem that party elites could
act as a brake on projects aiming to democratize political parties. At
the same time, it signaled that change in the parties and party system
would be more likely if the next generation of political leaders had
developed skills and motivations in favor of change. It is also an
example of how IPA donors do not expect immediate change: change
can happen with generational change. 21
Unequal gender representation has both symbolic and substantive
consequences for how democracies work. Severe gender imbalance in
political institutions diminishes the representative nature of such
institutions, and moreover, it leads to underrepresentation of political
interests. Significant efforts have therefore been made to increase the
number of women running for – and winning – office (Domingo &
et.al, 2012; Gouzou, Sanou, & Vedsted, 2014; UNDP/NDI, 2012)
(Kandawasvika-Nhundu, 2013; Roza, Llanos, & Roza, 2012). As table
2 displays the recruitment of women to political institutions (in this
case the lower house of parliament) shows consistent improvements
across all non-European regions.
21
However, KAS closed its office in Malawi in 2012 due to a strategic shift of attention
towards North Africa. The YPU project has received - particularly ahead of the 2014
elections - funding from NDI and the NICE programme (National Initiative for Civic
Education, formerly funded by the Eu and the German GTZ/GIZ).(Source:
Communication from KAS, 03.12.14)
29
Table 1 Women MPs as share (%) of total no. of members of lower house, by
geographic region (outside Europe)
Year*
Americas**
1997
12,6
2006
19,7
2014
25,6
Asia***
Sub-Saharan Africa
13,6
10,6
16,4
17,0
18,8
22,6
Pacific****
Arab States
10,8
3,6
12,3
8,2
13,4
17,8
* Year refers to info by 10.11.1997, 30.09.2006, 01.09.2014,**Incl. US and
Canada
*** Incl. Japan and India, ****Incl. Australia and New Zealand
Source: Compiled from figures available at http://ipu.org/wmn-e/world-arc.htm
(Accessed 01.10.14, NB Not all of the countries included in the table are new
democracies and therefore not recipients of IPA).
While it is possible that the many programs organized by donor
institutions have contributed to this development, there are also many
other factors that explain this change 22. Practically all donor
institutions involved in democracy assistance, work in this field and
multilateral organizations like the UN and IIDEA have had this as a
prioritized area. Gender issues have become integrated into the
development policy in general (OECD, 2012).
The increase in the number of women in parliaments is an example
of the interaction between various types of democracy assistance
programs. While some of the increase may have been the result of
changes in parties nominating more women candidates, in several
countries there has been a change in legislation which requires parties
to introduce quotas for women candidates in order to qualify to be on
the electoral ballot, or special seats have been introduced where only
women candidates may contest, such as in Uganda. Donor institutions
have also organized training for women candidates, whether belonging
parties or not (Muriaas, Tønnessen, & Wang, 2013). Economic and
educational opportunities, two variables known to impact political
recruitment, may also have improved for women and thereby
increased the number of female candidates. Thus, support for parties
alone cannot explain the increase in women’s representation, but
political parties in most countries is the main channel for recruitment
to political offices.
22
See for example (Irvine, 2013).
30
Improvement of women representation in politics is partly
dependent on political parties, but is as much dependent on changes in
the environment, as in legislation, and in improvement of sociostructural conditions. Thus, such objectives benefit from IPA
programs that look beyond parties themselves and relate the efforts to
other forms of democracy assistance and to development assistance in
general.
3.2
Organizational models for IPA
The implementation of IPA takes the form of a chain of delegation. In
the donor country there is usually one major financial source, such as
the country’s foreign office, ministry of developmental affairs or an
agency charged with handling such issues, such as Sida in Sweden. The
next link in the chain involves the units that organize IPA.
Donor countries have organized IPA along different models. 23
These vary between party affiliated institutions, as in Sweden and
Germany, and inter-party institutions as in The Netherlands, United
Kingdom, and Denmark.
Party affiliated institutions are organizational units established by
political parties to handle IPA, as in Sweden and in Germany. In the
German case these institutions perform research and information
activities, both domestically and internationally. Each of the major
parties has its own foundation. Originally they worked exclusively
with sister parties in recipient countries, but this is now prohibited.
Hence, they have turned to cross-party activities and to many projects
involving non-party actors 24. The two American institutions, IRI and
NDI are officially independent of the two respective parties in the US,
but widely seen as linked to either of them, such as by the
23
In this section I deal with nationally organized IPA. Other actors providing IPA is multilateral organizations, such as the International Institute for Democracy and Electoral
Assistance (IDEA) and the United Nations Development Program (UNDP). These
organizations work, by definition on a cross-party basis, and often develop handbooks,
information materials, organize seminars and provide analysis of topics that can be used by
parties (IIDEA, 2014b; Power & Coleman, 2011) (UNDP, 2005, 2006).
24
The Konrad Adenauer Foundation (Christian Democratic Union – CDU) Friedrich Ebert
Foundation (Social Democratic Party - SDP), Friedrich Naumann Foundation for Freedom
(Freedom Party – FPD), Hanns Seidel Foundation Christian Social Union of Bavaria –
CSU), Heinrich Böll Foundation ( the Alliance 90/The Green Party) and the Rosa
Luxemburg Foundation (The Left party).
31
composition of the board. Also these institutions are obliged to work
with several parties.
Inter-party institutions are organizational units established by acts
of parliaments, but with different ways of implementing party
assistance projects. The UK and Denmark for example has at the
national level a joint IPA unit for democracy assistance programs. The
Westminster Foundation for Democracy and the Danish Institute for
Parties and Democracy combine a sister-party approach with
multiparty projects. The implementation of the projects is partly
executed through the parties themselves, partly by the inter-party
institution. The now dissolved Norwegian Centre for Democracy was
also an inter-party institution, but the implementation of the project
was entirely controlled by the parties. The Dutch NIMD is quite
different from the other IPA donors in that it has a joint organization
at the national level which is also the main actor implementing the
IPA projects 25.
The reach of IPA institutions is obviously a function of its size,
combined with its geographical reach. The two American Institutions
claim to have, or to have had, offices in more than 100 countries, while
the two largest German ‘Stiftungen’ have a presence in more than 80
countries. NIMD is present in 20 countries.
In the recipient countries the operation of the IPA projects can be
executed through local offices of the donor organization, as is the case
with the German and American IPA organizations, or through a
locally organized NGO, as is often the case with the Dutch model, or
in partnership with other donors.
Staffing of the national offices varies between locally recruited
leaders, such as for NIMD, or nationals of the donor countries, as
practiced by the American and German IPAs. Many party affiliated
IPA donors lack a permanent representation in the recipient countries
but depend on ad hoc, or regular but infrequent, communications with
the recipient parties.
Although it cannot be proven that donors with country offices
achieve better results, there are several arguments in favor. A
continuous local presence obviously improves the possibility of
25
NIMD was originally closely connected to the parties in the Dutch parliament. The parties
had program officers working at NIMD and the board consisted of representatives for the
parties. NIMD is now organized separate from the parties.
32
continuously monitoring the implementation of programs, compared
to ad hoc presence, and adjustments can easier be made. A local
presence may also improve the exchange of information between
donors and facilitate the coordination with other IPA projects and
link IPA to other forms of democracy assistance. Party-to-party
projects can compensate for the lack of field offices by concentrating
their efforts geographically which would reduce the transaction costs.
3.3 Financial model and scale of intervention
The financial model of Sweden, Denmark and Britain has in common
that the funding for the party affiliated institutions is determined by
the electoral support that each of the parties receive in national
elections. The parties enjoy significant autonomy in choosing which
parties to cooperate with in new democracies. At least from a
resource perspective, IPA in these countries can in principle be disconnected from other development programs. This model helps
separate IPA projects from the official foreign policy of the donor
countries, but also risks de-coupling IPA from other types of
democracy support and from developmental efforts generally.
The scale of intervention refers to the amount of funds available
for IPA. Although there is no automatic link between the amount of
funds that are spent on IPA and the outcomes, most studies and
evaluation reports are critical of very small projects, particularly if
judged against the more macro-defined goal of improving democracy.
Funding for democracy assistance has increased in general, but it is
problematic to track the development of IPA for many donor
institutions. The scale of IPA is problematic to assess for four reasons.
First, donor organizations do not always provide publicly available
information about their financial resources. Secondly, as most IPA
donors also are involved in other types of democracy assistance, it is
problematic to distinguish IPA funding from other types of funding.
Third, publicly available data do not provide information about
individual IPA projects and finally, we often have no information
about how significant IPA finances are for the recipient parties. Party
finance is a closely guarded secret. It is therefore impossible to know
if IPA funding is so significant that it offers an incentive for parties to
change. This is particularly thru for incumbent parties which have
access to resources that the opposition is denied. For minor parties
33
IPA funding may be more critical because these will rarely attract
funding from wealthy donors.
Donor institutions that work almost exclusively with political
parties are the Swedish PAOs, the Danish DIPD and the Dutch
NIMD (table 2). In addition, WFD specifies the share of total funding
allocated to party assistance. Sweden is a major IPA provider
compared to DIPD and WFD, but slightly less than NIMD.
Table 2 Donor institutions and IPA funds (million EUR)
Sweden: PAO
Donors
Year(s)
2011-2014
Funding/ year
8.8
Denmark: DIPD
UK: WFD
2011-2013
2013-2014
3.4
2.6
2013
10.3
The Netherlands: NIMD
Sources: (Sida, 2014a:1), (DIPD, 2013:122): (WFD, 2013b:8): (NIMD, 2014a:18)
Sweden’s IPA program started in 1995 1.54 million Euro. For the
current 2011-2014 period slightly more than 8.6 million Euro are
available annually. The Danish program, which is not totally devoted
to party support, has available 3.4 million Euro annually 26. WFD
spends 2.6 million Euro on party related programs. NIMD works
almost exclusively with political parties. NIMDs total budget has
increased from ca.2.2 million Euro from its start in 2002 (NIMD,
2003:9) to more than 10 million Euro in 2013. The two largest
German donors, KAS and FES (Friedrich Ebert Stiftung), have
increased their total budget for international cooperation from
approximately 55 million Euro in 2006 to 60 million Euro in 2011, and
from 60 to75 million Euro, respectively (Weissenbach, 2014: 9-10)
But even if there has been a notable increase in funding for IPA,
many donor institutions have also increased the number of projects. It
is not possible in this report to estimate how much funds that are
allocated to individual projects. There are many projects that are cofinanced by several donors, although it is unclear how extensive such
cooperation is compared to the total funds available. Thus, although
26
In addition to these donor institutions Norwegian IPA funding for 2011 (the latest
published figures) was ca. 0.5 mill EUR, ca. half of the funding available to the dissolved
Norwegian Center for Democracy Support.
(http://www.norad.no/no/tilskudd/tildelinger/norads-tilskudd-til-politiske-partier-2011)
(Accessed 12.03.14).
34
funding for IPA has been increasing and the number of projects have
increased, IPA is small compared to development assistance in general
and to other targets for democracy assistance (Rakner & Svåsand,
2011:43).
The financing of Swedish PAOs has been based on the respective
parties’ share of the seats in the Riksdag, plus a basic grant irrespective
of the party size. In 2012 party assistance was more clearly
differentiated into a) support for sister parties and related political
organizations and movements (70% of the budget) and b) support for
multiparty projects (30% of the budget). The former is based on
program proposals for a three year period and funding for these are
based on the PAO-parties representation in the Riksdag. The latter
objective is supported based on competitive applications to SIDA
(SIPU, 2011).
A similar model is applied in Denmark where the budget is split
40% for bilateral projects for the parties and 40% for multiparty
projects. The former is run by the individual parties, while DIPD runs
the multiparty projects (DIPD, 2013). Of the funds for bilateral
projects, one third is shared equally between the parties represented in
the Folketing 27 and two thirds split according to the parties’ share of
the seats. The British WFD allocates almost half of the total budget to
the parties (£2.1 million), with ca. 41 % each for Labor and
Conservatives, 14% for the Liberals and the remainder (4%) for the
smaller parties.
3.4 Modes of intervention
Modes of intervention refer to the methods used by donors to
implement IPA. IPA projects cover many different modes. UNDP
distinguishes between direct and indirect support for parties. Direct
support targets parties as such, while indirect support is for
institutions or processes that parties also are part of or affected by.
Direct support, such as cash transfer to political parties for particular
projects is only one of several modes or the provision of services and
goods directly to political parties. The so-called ‘standard’ method of
IPA (Carothers, 2006: 113) is the organization of seminars or training
sessions, jointly for several parties or organized for one party at the
27
One of the eight parties did not apply for any IPA funds.
35
time. Several donor institutions also sponsor exchange visits to
neighboring states or to their home countries.
Indirect support is for example efforts to strengthen the parliament
and parliamentary groups, or the electoral process. Indirect support
has increased among IPA donors. This is because donors realize that
parties are strongly affected by the institutional environment.
Strengthening the parliament, including the party groups, is seen as a
way to improve horizontal accountability in the political system and
to improve parliament’s oversight functions. Improving the
parliamentary arm of the parties without simultaneously improving
the quality of the extra-parliamentary party organization, may on the
other hand tilt the power balance internally in the parties more in the
favor of the elites. Strengthening parliament is also an argument for
why IPA should be closer linked to other forms of democracy
assistance.
A critical issue in the provision of direct party support is the
question of what relevance a particular type of support has for the
objective of IPA. Relevance cannot be defined in the abstract,
detached from the recipient parties. Thus, a key element in IPA is that
the recipient parties see the projects as relevant and acceptable.
Therefore trust among donors and parties is one of the most
important factors influencing whether an IPA project works or not.
Both the literature on IPA and the evaluation reports document
frequent examples of failed projects because of resistance to projects
among the key personnel in the recipient parties. Although many
projects are relevant, it does not mean they have impact. An example
of relevant projects could be seminars to train party officials in
organizational routines. One could easily document the outputs of
such projects: how much that was spent and how many that
participated in seminars. It is more problematic to document the
impact; strengthening party organization. While the individuals who
participate in seminars could benefit it does not follow that the party
organization would change in the desired direction.
The German foundations and the American institutes cover much
more than party assistance. NIMD has chosen a rather unique
approach for party and party system assistance. In most countries
where NIMD is operating it has established a locally organized NGO
where parties with parliamentary representation are members, usually
referred to as Center for Multiparty Democracy (CMD). Only parties
that participate in the CMD are eligible to apply for project funding
36
directly from NIMD. This is an attempt to link two of the objectives
for NIMD: improving inter-party dialogue and improving party
organizations.
3.5 Timing and endurance of IPA projects
IPA is meant to contribute to the development of a democratic
political system. As we have seen some IPA programs are motivated to
bring an end to an authoritarian regime, while others seek to influence
the transition process itself. But most IPA programs are introduced
after the transition in order to contribute to democratic consolidation.
IPA programs prior to the transition to a (more) democratic state
in a country often have electoral victory for opposition groups or
parties as a main objective in the short run, while IPA programs
running after a transition phase are more concerned with party system
and party institutionalization. Party system institutionalization and
individual party institutionalization take time. There is no clear end
point for institutionalization and institutionalization can also be
reversed. The longevity of IPA programs is therefore vital. IPA
programs differ therefore widely in both when they start and how
long they last. A frequent critique in evaluation reports is that many
IPA projects have a short time-horizon and are without clear ideas
about follow-up projects. Because institutionalization is a long-term
process, one can expect that IPAs can have a stronger impact if there
is continuity, rather than if projects are organized ad hoc.
Once a donor has started IPA in a country, there are strong
incentives to continue. There are ‘sunk costs’ and if at least some
progress has been made, everything could be lost if the actor
withdraws 28. Moreover, it takes time for actors, individual parties as
well as multi-party institutions, to develop contacts, establish trust
with recipient parties and accumulate knowledge of the cases. There
are therefore good reasons for the continuity of projects even if it is
not possible to identify clear progress in the short run. Although
NIMD, for example, has been active in Mozambique since 2002, it can
be argued that not much have changed, either in terms of individual
parties or the party system as such. Mozambique continues to be a
28
See (Brown, 2011) for a discussion of democracy promotion arguments used among donor
institutions in spite of the absence of visible progress.
37
dominant party system and neither the quality of elections or the state
of opposition party organizations seems, to some observers, not to
have improved very much (Manning, 2010; Sumich, 2010).
When an IPA donor exits from countries it is rarely because it
concludes to have reached the objectives. NIMD terminated its
programs in Tanzania and in Malawi because of the change in Dutch
development aid policy, but the programs were continued by other
institutions, such as the UNDP and DIPD. Likewise, FES, the
foundation of the German Social Democratic party, shifted its focus
towards the former Communist countries of Eastern and Central
Europe and the Balkans as Africa “….retreated from the view of the
Social democratic party and the leadership of FES” (Vinnai, 2007: 89).
Similarly, the United States scaled back its democracy support
(including party support) in Central and Eastern Europe when EUmembership became a possibility for countries in that region and it
was assumed that EU donors would replace the American donors. The
need for IPA has also been reduced in many of the new democracies in
this region as parties have come under the umbrella of EUparliamentary groups and European party federations, but also because
they have introduced public subsidies for parties (IIDEA, 2014a: Ch.
8).
In some cases a donor has abandoned programs because of
problems with project implementation. In Zambia in 2010 and in
Bolivia in 2013 NIMD ceased its activities because of problems with
its local partner institutions (NIMD, 2011, 2014b).
The time dimension of IPA is crucial. All documents published by
donor organizations, whether multilateral or national, emphasize that
both party institutionalization and party system consolidation are not
achieved overnight. In light of this argument it is not surprising that it
has been difficult to demonstrate the effects of IPA.
3.6 Geographic focus
The great expansion in democracy assistance in general, and in IPA,
followed in the wake of the Third wave of democratization
(Huntington, 1991), and particularly after the fall of the Berlin wall.
Hence, with the exception of the long-running German party
foundations, most IPA programs were focused on the former
38
Communist regimes and East and Central Europe and the states
emerging from the dissolution of the Soviet Union.
The two largest German foundations, KAS and FES have ca. 80
field offices and work in more than 100 countries, with FNS, the
Liberal party foundation, being represented in far fewer countries.
(Involvement in IPA is only one of their activities). Likewise, the two
American democracy assisting institutions have a world-wide spread.
Table 3 No. of new democracies 2014*with representation for the three largest
German foundations, the two US foundations, NIMD, WFD and project
countries for Swedish PAO
Donor
institution
KAS
FES
FNS
NDI
IRI
NIMD
WFD**
Swedish
PAO
Africa
10
19
5
20
9
9
4
24
LatinAmerica
Asia
12
20
2
8
9
5
11
19
10
13
11
2
3
1
8
Eurasia
2
8
9
9
6
1
1
6
Middle East
6
10
10
16
8
1
4
5
Balkan
5
7
7
5
3
-
1
5
Acronyms: See appendix. Sources: The foundations’ web-sites. (Not all web-sites
are updated and no. of countries may therefore deviate from the actual no. of
countries with programs currently). Excluding offices in established democracies
and regional programs. For Swedish PAOs: Dokument Sida/INTEM DEMO (JN)
2014-04-05 (not all country programs shown in the document).
** No. of countries refers to WFD operated programs, not those by individual
parties.
Compared to the ‘big five’, NIMD and WFD which both work with
multiple parties are smaller and have a greater concentration on Africa.
Most projects such as the Swedish PAOs, where targeting parties in
Central and Eastern Europe in the beginning of the programs. In the
early years of the program 4/5 of the funding went to projects in
Eastern and Central Europe (with the Baltic states topping the list of
recipient countries), with 1/3 going to developing countries, but
shifted more towards African countries.
The change in geographic location is partly a result of increased
funding for IPA so that donor institutions can spread their projects.
However, it also reflects changing opportunities and needs. As the
wave of democratization led to the end of one-party states or military
39
rule in Africa, Asia and Latin-America, donors turned their attention
to new democracies in these regions. When many of the former
Communist-ruled countries gained European Union membership
there was a further shift towards non-European countries.
As most bilateral IPA programs depend either on the official
development policy of the donor countries, as for example is the case
for NIMD, or on the preferences of the individual parties, it means
that in some countries there can be several donors active at the same
time, while in other countries there is only one donor, or one donor
dominates. Even when there are several donors in a country, most
donors operate their own projects, although they are often informed
of each other’s work. It is a big unknown in the study of IPA what
the effects are of several IPA projects operating simultaneously, but
uncoordinated. It is possible that the cumulative effects will
contribute to the overall objectives, but it also possible that projects
will pull in different directions.
3.7 Summary
The dimensions of IPA that relate to the objectives, organization and
modes of intervention depict clear contrasts between donor
institutions and across time.
Sweden in particular, together with Denmark and the UK, is
leaving much more of the IPA field to individual parties. This
concerns both the funding mode and with whom to cooperate. NIMD
represents almost the opposite. Although some Dutch parties have
foundations for IPA, the dominant actor in the Netherlands is NIMD.
The choice of countries is closely linked to Dutch development policy
and the mode of intervention is inclusive as regards the parties. But
NIMD share with the other countries a strong focus on parties as
such.
In between these positions are the German foundations and
American institutes. They share with NIMD the inclusion of multiple
parties, but they differ from NIMD in having several other targets for
their work. Across time, there is a clear shift in democracy assistance
away from a party-to-party approach and concentration on parties per
se, towards the institutional environment of parties and towards the
rule of law.
40
There are pros and cons of multiparty approaches and party-toparty approaches. If the objective is to assist in democratic
development, then multiparty approaches, including inter-party
approaches, seem more likely to contribute to this objective than
party-to-party approaches. Multi-party approaches avoid the potential
problem of taking sides in internal political affairs. Party-to-party
program can, however, contribute to the institutionalization of the
recipient parties.
41
4. Assessing the effects of IPA
Studies of IPA often raise the question: Does IPA have effects that
correspond to its motivation? To answer this question, we must first
address some methodological issues.
4.1 Challenges in assessing the effects of IPA
IPA is an instrument used in pursuit of several objectives. IPA actors
work in different kinds of political systems, at different stages of
democratic transition and consolidation, and for shorter or longer
time periods. IPA programs differ also in terms of approaches and
with different budgets. Because of the variations among IPA programs
they are likely to have diverse effects. It is anyway recognized that it is
hard to prove what the impacts of IPA are. The problem of analyzing
effects of party assistance is primarily methodological, related to the
availability of adequate information. In general, the purposes of IPA
are to promote change in a desired direction; such as towards more
stable party systems or towards more institutionalized political
parties. These changes are assumed to contribute to democratic
consolidation. The forms and the scope of IPA are independent
variables that are assumed to have a positive impact on one or more
dependent variables (stable party system or institutionalized political
parties)
In general, to speak of a cause having a particular effect requires
that several conditions are fulfilled: 1) there must be a mechanism
linking a variable (for example modes of party support) and the
outcome, 2) the cause (party support) must come before the observed
change in the dependent variable, 3) the two variables must have covariation. We must be able to observe that change in independent
variable (cause) corresponds to observed changes in dependent
variable (effect). 4) Alternative explanations must be accounted for
and rejected.
Two caveats are relevant to any findings of effects. First, a ‘cause’
can have an effect as aimed for, but also unintended effects, or in the
worst of circumstance, primarily unintended effects. An example
would be if the prospect of international party support triggered the
formation of new parties, rather than the consolidation of existing
ones. Second, observed effects may be caused by other factors than
42
IPA, such as implementation of legislation on party finance or
regulation on the defection of parliamentarians from one party to
another. In addition, even if IPA can be credited with having a desired
effect, the relationship may be contingent on specific circumstances.
IPA is one of several ways democracy assistance is being provided,
including
support
for
parliaments,
electoral
processes,
decentralization, and media and civil society. Political parties are
institutions that are connected to, or impacted by, all of these other
targets for democracy assistance. Thus it may not be possible to
isolate the effects of IPA from changes in other institutions which
interact with political parties 29.
In addition, in some countries there is only one donor, or one
dominant donor, while in other countries there are several donor
institutions each with many small projects. Each individual IPA
project may be too small to have an impact on macro-level
developments, but where there are several donor institutions it is
possible that the IPA projects cumulatively have contributed to
stronger parties, even though it may not be possible to specify the
contribution of each individual donor organization.
Moreover, even if it is possible to conclude that the causal variable,
IPA, has had a positive effect on party developments in a particular
country, it does not follow that similar effects will be found when the
same way of implementing IPA is transferred to other contexts, or
used in the same way over longer time periods. The call for context
sensitivity is repeated throughout the literature on democracy
assistance, including IPA. Context refers not only to geography, but
also to the time dimension within each geographic context. Thus, an
effective way of organizing IPA prior to a transition from an
authoritarian regime to a democratic regime need not be similarly
relevant once a new regime is being established.
29
The impact of party assistance cannot for instance be gauged by field experimental
methods using randomization. This methodology has been increasingly applied in studies of
developmental assistance in general, but also in the study of democracy and governance
assistance programs (Moehler, 2010)
The reason why this method has not been suitable for the study of party assistance is that
randomization requires a high number of units of analysis and is therefore primarily applied
in studies of lower-level units, such villages and municipalities. However, political parties are
institutions functioning at the macro-level. Although it is conceivable that randomized field
experiments could be used to study variation within a given political party, the party system
cannot be disaggregated into sub-national units
43
It is also a challenge in all studies to have a clear specification of
what the ‘causal variable’ is supposed to change. There are at least two
issues to consider here. First, to reiterate: the overarching goal of IPA
is to contribute to democratic consolidation, or improvement of
democracy. Democracy is a multi-dimensional concept, but primarily
a macro-level characteristic; a quality of the political system as such.
The mechanisms of IPA target political parties, which are units
functioning between individuals and the political system, so-called
meso-level units. Thus, through IPA one hopes that by changing the
characteristics of the meso-level units, improvements at the macrolevel will follow30. But, as discussed initially, change towards more, or
better, democracy at the macro-level depends on multiple factors.
Second, attempts to measure the impact of IPA require data
observations for at least two time points; prior to and after the
introduction of the causal variable. Data that are needed vary with
what the objectives of IPA are. For example, if the objective is to
contribute to the consolidation of the party system, there are a
number indices available that capture key indicators of party system
characteristics; such as party fragmentation, electoral volatility or the
extent of party representation in the parliaments (Lijphart, 1994).
Party system characteristics are generated from official electoral data
at the macro-level. While such indicators do not tell the whole story
of party system consolidation, they do provide some means of
assessing change over time in what is usually considered to be
important qualities of party systems which may affect the quality or
stability of democracies (Russel J. Dalton, 2008; Wang, 2014;
Xezonakis, 2012).
Assessments of whether IPA contributes or not to
democratization in general must take into account both the particular
empirical context in which IPA functions as well as the institutional
environment of IPA programs. Support for political parties in
authoritarian countries cannot be structured and implemented in the
same way as in in countries where civil liberties like freedom of the
press and freedom of association are generally respected. When the
electoral playing field is strongly tilted in favor of the incumbent
party, the functioning of the parties during the election campaign will
30
The relationship between the qualities of political parties and the qualities of democracies
has been a controversial point in the party literature since the early 1900’s (Michels, 1965;
Ostrogorski, 1964; Schattschneider, 1942). See also (Assarson, 1993; Teorell, 1999) for a
discussion of whether parties should be internally democratic.
44
obviously be different than in a system with fair and free elections. In
countries with a dominant incumbent party the needs of the
opposition party/parties will be different than in a system of
competitive parties.
There are few agreed upon indicators regarding party
organizations, but several proposals identifying key characteristics of
party organizations have been developed (Basedau & Stroh, 2008;
NIMD, 2004b; Poguntke, Obert, & Tipei, 2013). The problem
however, is primarily access to information that can be used. For
example, membership figures are often considered a key indicator of
the strength of party organization. Yet, in many new democracies
membership figures are not available, or are considered unreliable.
Another indicator is party finance, where again data are often not
available or not trustworthy. Therefore, assessment of change in the
state of party organizations is much more problematic than change in
party systems.
The time perspective for detecting changes caused by the
independent variable complicates the study of effects. It may not be
possible to see any effect of an independent variable until after several
years of support. International party assistance is recognized to be of
this kind31. Even if a change can be observed after a short time, there
is no way of knowing if the effect is lasting or fades out in the long
run. Party institutionalization does not have an end-point from which
it cannot be reversed.
In general, a big problem in assessing the impact of IPA is the
availability of information on IPA projects: who gets how much, how
long, in what way and for what? The public information that is
available on the donors’ websites and in published reports covers, at
best, the total amount of funds budgeted for a country. However, this
includes usually also the funds needed to operate the donor
organisation’s branch in the country, cross-party activities and
activities not specifically targeting political parties. The financing of
individual party projects is usually not available. Moreover, it is
impossible to know how significant IPA programs are for the
recipient parties as party finance is a well-kept secret.
31
«Political party assistance can take years to bear fruit» (UNDP, 2006:80)
45
The total amount of democracy assistance is miniscule compared to
ODA in general. It is therefore needed to have realistic expectations
to what IPA projects individually and collectively can achieve in the
short run. In general it is a big unknown what the financial
significance of IPA is for the individual parties. For example, the
ANC in South Africa and the NRM-O in Uganda have been in office
for a very long time, two decades in the case of ANC, almost three
decades in the case of NRM-O. The two also have extensive
organizational resources, while their opponents often lack resources.
For these two parties participation in IPA programs may be important
politically, but it is not likely that it matters very much economically.
Thus, the leverage that donors can exert is probably minimal in such
cases. On the other hand, for smaller parties, even small projects can
be vital to keep the organization alive and to institutionalize. Whether
this is an objective for IPA is a different matter.
4.2. Does IPA work or not?
In order to be able to answer the question if IPA works or not, a
considerable amount of data must be available. Ideally, we should have
data across time and across countries. We should have data on
individual IPA projects as well as macro-level data comprising all IPA
efforts in a country. It is beyond the scope of this paper to conduct
such an empirical investigation, but as a first step I have tried to
summarize information from research literature and evaluations from
some of the IPA donor institutions.
Research on IPA
There is a growing research literature dealing with international party
assistance, but highly varied in terms of focus and methodology, as
well as in terms of time perspective, timing and geographic focus.
Some studies are focusing on individual countries while others have a
comparative approach. Tom Carother’s volume “Confronting the
Weakest Link. Aiding Political Parties in New Democracies”
(Carothers, 2006) is the most comprehensive study of IPA. The
essence of his study has been summarized in the formulation that “It
rarely has transformative impact, despite the hopes and sometimes
belief of its providers” (Carothers, 2006:160). The reasons for this is
46
that “the main determinants of the make-up and methods of parties in
the recipient countries are an array of underlying economic, political,
social, and cultural conditions that are largely beyond its ambit”
(Carothers, 2006: 214-215). Transformative effects are, for Carothers
when recipient parties “substantially ameliorate the central
organizational or functional characteristics that the aid providers seek
to address”(Carothers, 2006: 163). An example of this would be if
leadership fixated party without a grass-root organization was
transformed into a membership based organization with internal
democracy. Carothers find that there are some examples of change
due to IPA, most notably in campaign methods There are also cases of
small organizational changes, as when a party that has not held a
national convention actually do so. “The hard question is whether the
small positive steps will deepen over time and cumulate in major
organizational changes” (Carothers, 2006: 187). There are more
examples of projects that have failed to achieve their objectives.
Attempts to induce parties to change are met with opposition from
party leaders who fear they will lose control of the parties. In other
cases the environment, both the institutional (for example the role of
the monarch in a polity) or economic (such as wide spread poverty)
works against efforts to create well- structured parties.
The edited volume by Burnell and Gerritt (2010) is the most
recent compilation of research on IPA. A single source summarizing
the effects of IPA has not been made since Carothers’ book. The
articles cover assistance programs in the Balkans, Central Asia, LatinAmerica, countries in the Middle East and in Sub-Saharan Africa. But
the empirical material for these studies is very diverse. Some of these
articles conclude that IPA does not work, or that IPA has unintended
effects, while others argue that there are some examples of positive
outcomes.
In sum, the main finding from this research is largely confirming
Carothers study from 2006, and some are very critical as is evident in
Baader’s study “Party politics in Georgia and Ukraine and the failure
of Western assistance” (Bader, 2010. But in two areas it is argued that
IPA has made a difference. The first is IPA efforts to contribute to
regime change. Studies of Serbia identify IPA projects as contributing
to the defeat of Milosovic regime and support for ANC in South
Africa contributed to overturning the Apartheid regime as well
providing strong support for the new democracy. A second area is in
the recruitment of women into politics in Croatia and Serbia Here, it is
47
argued that IPA projects were a major factor in improving women
candidates’ electoral chances.
But no study is able to argue that IPA has had a transformative
effect; that is changing the fundamental structure of the party
organizations or the party system. Studies from Zambia and Malawi,
for example conclude that relationship between political parties has
improved as the parties with representation in parliament agreed to
meet regularly in the inter-party dialogue forum, which was a
requirement in order to apply for support for individual party
projects. However, such projects do not seem to have strengthened
party organizations and party fragmentation has continued. This
finding echoes that of studies of Morocco and Kenya where it is
difficult to find indications of fundamental party change. A common
explanation for why IPA has not succeeded better is that the
established elites are not interested in party reform. There are also
institutional factors that restrain the impact of IPA, such as role of the
king in Morocco, and the structure of the electoral system in BosniaHerzegovina which has reinforced the parties’ mobilization of
particular ethnic groups while the ambition of the donors have been
multi-ethnic parties. Several of the studies blame the IPA donors for
failing to grasp the essence of the local context, as much as the
recipient parties.
This research literature is nevertheless very scattered. Few
published studies have followed IPA projects over longer periods, and
in most cases, the studies focus on the efforts of one particular donor,
but not the total IPA funding that has been made available in a
country.
For more detailed information on research literature see Annex A.
Evaluations of IPA
As indicated in the introduction to this report, a main source of
information on IPA is available the evaluation reports commissioned
by donor institutions, but the use of and structure of evaluation
reports differs a lot between the donors.
Some reports examine a single donor institution in a particular
country, for example (Dijkstra & Kumado, 2004) which studied
NIMD’s program in Ghana for the years 2000 to 2003. In contrast,
48
the American National Research Council’s report (NRC, 2007)
covered the whole field of democracy assistance supported by the
National Endowment for Democracy (NED), where party support is
one of several activities, The first two evaluation reports of the
Swedish PAO (Uggla et al., 2000; Öhman, Öberg, Holmström,
Wockelberg, & Åberg, 2004) paid a lot of attention to the
organization of IPA in Sweden, whereas the third evaluation included
three country cases.
Another challenge with using evaluation reports relates to timing
and the fact that some of them are conducted shortly after a program
has been started. An example is the evaluation of the Danish Institute
for Parties and Democracy (DIPD). The institute was organized in
2011, but already two years after an evaluation report was
commissioned (MFA, 2013). The report therefore has more to say
about the organization of DIPD than about the impacts of the party
Evaluation reports rely on several sources of information. The
point of departure is usually the policy documents of the donor
institution, such as plans for IPA, decisions about goal formulation,
budgetary data, implementation plans and monitoring and evaluation
routines. Documentary evidence is supplemented with interviews with
staff members, both at the donor headquarter and in the field. Most
evaluation reports also include short term field visits to a few of the
countries where the donor institution has been active. A prime
purpose of the field visits is to collect information from the recipients
of IPA through interviews with actors in the political parties, but
often with other actors as well, such electoral management bodies and
civil society organizations working on elections. But apart from
interviews with recipient party representatives evaluation teams rarely
collect data independently. Evaluations are carried out by different
types of experts. It is common to engage independent companies that
specialize in evaluations, but they may not necessarily be country
specialist. Often, academics who are either area specialists or who have
analysed relevant topics (elections, parties, democracy) are included in
evaluation teams. All of these aspects of evaluation reports obviously
affect the comparability of their findings.
IPA projects vary along several dimensions: geographic focus,
objectives, organizational model, financial resources, modes of
intervention, and the timing of intervention. (Of these dimensions,
there is less systematic data on financial resources on the project level,
than for the other dimensions)
49
The first thing to note about the evaluation reports, even for the
same IPA actor, is the variation in success rates. Projects that are
found have succeeded or failed are scattered across many countries. It
is the specific country context, more than geography as such that is
important. However, it is possible to detect a regime effect on IPA
projects. In states that are semi-authoritarian, as in the Post-Soviet
Caucasus states, efforts to strengthen parties have largely failed and
programs have been re-oriented towards civil society or to
environmental factors, such as the media structure. IPA efforts have
also largely ceased in Eastern European states after these states gained
membership in the European Union which exerts strong influence on
the party systems.
The objectives of IPA have partly been to contribute to democratic
transitions, partly to improve individual parties and partly to
contribute towards stabilization of the party system. The first
objective was met in cases like Serbia, where NDI and IRIR supported
the opposition movements, including civil society organizations. But
most IPA projects are focused on democratic consolidation. The main
finding from evaluation reports is that the organizational model of the
IPA actor does not appear to be consistently related to successful vs.
unsuccessful outcomes of projects. Regardless of whether or not an
IPA actor operates along the same model as the Swedish PAOs or as
the Dutch NIMD, one finds some cases of success and others where
the projects cannot show any significant results. For example, some
projects by Swedish PAO’s in Colombia to strengthen individual
parties were found to have succeeded in their objectives, while other
similar projects by other PAOs, also in Colombia, failed in doing so.
A key finding in the evaluation reports is the enormous variations in
‘success’ when IPA projects are organized by individual political
parties in the donor countries. Naturally, individual political parties
vary in their capacity to organize and implement projects. In general,
the evaluation reports for WFD, DIDP and the Swedish PAOs find
that party-to-party projects do not have party system effects, which is
not surprising given the organizational model, but there nevertheless
is a gap between the overall objectives of democracy assistance and its
implementation.
The Dutch NIMD objectives are partly to improve inter-party
dialogue and partly to strengthen individual parties. Inter-party
dialogue has been the most successful element of NIMD’s programs.
But there is more variation in efforts to strengthen the party
50
organizations. In Ghana and in Mali, evaluations found positive
outcomes in both objectives, but in Zambia, Tanzania and Malawi, it is
difficult to detect strengthened individual parties. Efforts to stabilize
the party system, or efforts to improve competitiveness in dominant
party systems have been less successful, for NIMD as well as for all
other IPA actors.
A common experience across IPA donors is that where donors
meet resistance against change among incumbent party elites, efforts
to strengthen party organizations often fail. In some countries IPA
seems to have succeeded better for several parties, such as in Ghana
and in Romania, where the socio-economic environmental conditions
are more favorable compared to some of the surrounding states.
Another finding is that donors working on democracy assistance in
general (NDI, IRI, WFD), rather than with parties specifically, are
better situated to adjust their projects towards other objectives, such
as working with civil society organizations or focusing on
improvements in the environments of the parties, such parliaments
and the media structure.
All evaluation reports also questions the sustainability of what has
been achieved and call for more continuous projects, rather than single
events and for follow-up activities.
The evaluation reports arrive at many of the same conclusions as
the research on IPA. There are some cases of successful projects at the
party level and some improvements of the inter-party relationship.
However, fundamental change cannot be detected. The environmental
factors, both in terms of the political institutions with which parties
interact, and the resource availability, impact on the chances for IPA
projects to succeed
For more detailed information on donor evaluations, see Annex B.
51
5. Concluding reflections
Based on the literature on parties and on evaluation reports of IPA
programs it is hard to draw conclusions about their effects. There are
many examples of positive outcomes. Failures may occur because of
the way the projects have been designed or implemented, but could
also be a consequence of the environment in which projects are
implemented. It is also necessary to be realistic about what IPA can be
expected to achieve, given the scarce resources that are allocated to
IPA and the need for a long-term perspective. Carothers formulates
this quite succinctly: “The effects of party aid also tend to be quite
limited because the main determinants of the make-up and methods of
parties in the recipient countries are an array of underlying economic,
political, social, and cultural conditions that are largely beyond its
ambit”. (Carothers, 2006: 214-215).
5.1 A new IPA agenda?
One can notice a shift in emphasis over time in IPA. IPA has been
criticized for attempting to export a party model that does not any
longer exist, even in established democracies. It has also become clear
that strengthening parties and party system is strongly impacted by
the environment which comprises both legal and institutional
environments, as well as the socio-structural context.
More recent policy documents by IPA donors are less concerned
with party organizations than with party functions, particularly policy
development. Programmatic parties provide the electorate with
different policy packages. Donors have been concerned with the lack
of programmatic parties for several reasons. Parties in new
democracies have been seen as dominated by individuals rather than
with ideological perspectives, and elections have been fought between
individuals rather than between policy alternatives. Another problem
has been that parties have mobilized particular ethnic and/or religious
groups, rather than having an agenda for the nation as a whole. Both
donor institutions (NIMD, 2012) and policy analysts argue that
programmatic parties increase the quality of the democratic process
(Gonzalez-Acosta, 2009; Keefer, 2011, 2013). Delivery, or failure to
deliver on electoral promises, allows voters at future elections to hold
their representatives to account. Programmatic parties are focusing on
52
collective goods, either for the whole electorate or for larger groups in
the electorate.
In contrast, in the absence of programmatic parties, elections are
dominated by clientelistic appeals to smaller groups of voters which
are ‘rewarded’ , but the political system does not benefit from more
comprehensive and developmental policies (Keefer, 2005, 2011).
Clientelistic practices preserve the status quo, while according to
Keefer “… preliminary evidence indicates that parties able to convey a
programmatic policy stance or to discipline party leaders are
associated with significantly different policy choices across countries,
in both democracies and non-democracies” (Keefer, 2011:3).
However, programmatic orientation should not become a
straightjacket blocking compromises between parties. In many
political systems there is a need for either coalition governments or
compromises between institutions, such as is the case when one party
holds the presidential office without having a majority of the seats in
parliament. ‘Extreme’ commitments to programmatic orientations
may in such cases lead to gridlock, or in the worst case to democratic
breakdown 32.
Thus for international donors improving the policy-making
capacity of political parties is only one of several ways parties can be
expected to function better. It is expected that its programmatic
orientation will have a positive impact on social and political
development (IIDEA, 2014b), but this is also likely to be impacted by
the degree of trust between parties. The factors that impede
programmatic orientation are to be found, partly in the political
parties themselves, partly in the environment of the parties. To the
extent that the factors the IIDEA report and the articles by Keefer
et.al point to indeed are the ones that influence the development of
programmatic parties, international party assistance can contribute to
changing some of these factors, while others are clearly in the
‘periphery’ of what we can expect international party assistance to
change.
32
See also Stokes for why in some cases elected office holders defect from their
programmatic commitments: “Politicians who switch (policies) are representative; within
the bounds of what they know and believe, they are attempting to serve the interests of
citizens as best they can. Because they anticipate that voters’ beliefs are inaccurate, and
hence the their (induced) preferences over policies will change, politicians violate mandate
responsiveness in order to act as good representatives” (Susan C. Stokes, 1999: 103) (Ithalics
in the original).
53
It seems likely that some coordination of IPA with other
democracy assistance programs would enhance also the policy
development of political parties. For example, for parties to pursue
their electoral platform or to coalesce with other parties, it will be an
advantage if parliamentary procedures allowed standing committees
with adequate time to scrutinize governmental proposals and available
resources for parliamentarians to exercise their control function. In
such a case, support for parliamentary reform could go hand in hand
with support for political parties (Burnell, 2009).
5.2 Should IPA be maintained?
I have only briefly referred to some of the normative issues
concerning IPA, but I believe it is necessary to reflect on why IPA
should be maintained or not, particularly as it is difficult to prove that
large IPA programs that involve millions of Euros actually contribute
to complex objectives such as institutionalization of the party system,
and ultimately of democratic consolidation. There are important
arguments against IPA. IPA involves external actors more or less
directly in political processes that are essentially domestic affairs. The
development of what of kind of political parties, how they look like,
which political objectives they have and who the actors should be are
important elements of sovereign states. 33 Moreover, as has been
shown in research and evaluations of IPA it is sometimes unclear what
it is that IPA seeks to achieve and under what circumstances IPA is
successful. If it cannot be demonstrated that IPA works, why should
it be pursued?
Nevertheless, in spite of the lack of ‘hard evidence’ of the effects of
IPA programs, there are also some arguments that speak in favor of
maintaining (and expanding) IPA.
First, there are no democratic political systems that operate
without political parties. This observation is not an argument in itself
for IPA, but invites a reflection of why it is that modern democratic
polities have parties. The argument for the continuation of IPA is that
in democracies no other institution combines the functions of
representation, formation of government and opposition and interest
33
See (Agne, 2014) for a discussion of the relationship between democracy promotion vs.
how democracy is defined.
54
aggregation. All large scale communities must delegate to a smaller
group of representatives the authority to make decisions. The party
label of candidates seeking to be elected as representatives can tell the
voters something of the expected behavior of representatives. It is
important to underline can, as there are a number of conditions that
apply for the party label to provide that information. However, in the
absence of the party label, there is no way voters can be informed
about the political views of all competing candidates unless devoting
an extraordinary amount of time and resources.
The second function is related to the first. Once elected, unless
there are political parties, the establishment of groups supporting vs.
opposing the government will have to be constructed ad hoc, or
informally. Again, this undermines the electorate’s information about
what political alternatives there are to the current government. The
lack of an institutionalized party system prevents voters from holding
groups of representatives to account and limit the electorate’s
possibility of making informed choices for the next election.
Finally, elected representatives have the responsibility of taking a
holistic view of policy making, so-called interest aggregation. And this
point is perhaps one of the strongest arguments for why IPA should
be maintained.
Democracy assistance has for many years focused on support for
civil society organizations (CSO) – which are also perceived to be
central to democracies. CSOs can act as ‘watch-dogs’ towards political
authorities. But the support for CSO can also have side-effects that
are not beneficial for democratic developments 34. On the one hand
CSO support can lead to a mushrooming of CSOs which can although
benign, also result in increased demands on the political system. In the
worst case there can be organizations that do not really have a base in
the society but are dominated by a leader. CSOs that are heavily
dependent foreign funding may become more accountable to external
actors than to their members 35. More importantly, however, is that
there may be a (potentially) unintended consequence of a strong civil
society: an imbalance between the capacity in society to raise demands
vs the ability of the political system to prioritize among the competing
34
See (Hahn-Fuhr & Worschech, 2014) for a very critical assessment of civil society
promotion to Post-Soviet States and (Edwards & Hulme, 1996) for a similar cautioning
concerning NGO support in general.
35
This can also be a problem for political parties lacking domestic financial resources.
55
demands. CSOs are not accountable to the electorate, when they have
members and if they have internal democratic procedures. CSOs are
also concerned with segments of public policy – environmental
protection, promotion of religion, care for vulnerable groups,
provision of health and social services, or primarily for the
advancement of group interests, such as is the case for unions or
farmers. But CSOs have no responsibilities for balancing various
interests against each other and for the distribution of scarce public
resources. This is the essence of interest aggregation and policy
choices: only political parties offer a collective instrument for this. It
may therefore be argued that it is necessary to strengthen political
parties as a counter-weight to a strong civil society, but also that
political parties and civil society together can lead to more robust
democracy in which elites and citizens are linked through multiple
channels.
There is also an additional argument for why Sweden, and other
smaller democracies, should maintain IPA programs. As discussed
earlier in the report, the goals of IPA have been defined by political
authorities and when looking at the history and scope of IPA,
democracy assistance has partly been tied to assist a transition from
authoritarian to democratic regimes, partly to the consolidation once
the transition has taken place. Smaller democracies, like the
Netherlands and the Nordic countries, represent a different way of
organizing democratic governments than some of the larger countries
involved in IPA. Knowledge about the functioning and structuring of
the Nordic polities, including inter-party relationships, and the
difficult art of coalition-building, can provide a counter-weight to the
party system models of some of the larger IPA donor countries. This
is not to argue that the Nordic ‘model’, is necessarily better suited to
assist democratic consolidation in all cases. Nordic party politics
function within an institutional setting of proportional representation,
parliamentarism, and extensive local government, while many new
democracies have presidential systems and operate with majority
elections and limited local government. But the consensual model of
politics frequently associated with Nordic and Dutch politics
emphasize cooperation, inclusiveness and compromise as means to
achieve political stability.
56
5.3 If IPA is maintained: what should be done?
There is a need to be clear about what IPA is supposed to contribute
to. IPA is usually part of the overarching goal of assisting a process of
democratic consolidation. A sub-goal of this is often formulated as
assisting the development of a functioning party system. The
overarching goal calls for coordination of IPA programs with other
democracy assisting programs as democratic consolidation depends on
multiple institutions and processes. The sub-goal, a functioning party
system, calls for a cross-party, or an inter-party orientation. Bilateral
programs, where parties in donor countries assist selected (sister)
parties in recipient countries can contribute to the overarching goals,
but will not by automatically do so. But there are also positive impacts
of bilateral programs. Parties that are being harassed by authorities can
gain international support, and therefore survive. Parties without
sufficient domestic resources can institutionalize with assistance.
Party-to-party support may also increase the legitimacy of IPA in
general in the donor countries. The main issue, however, remains: how
can it be argued that bilateral programs are indeed linked to the wider
goals of party system consolidation and to the overarching goal of
democratic development?
Just like ‘democratic consolidation’ does not have an end-point,
party and party system institutionalization are continuous processes.
IPA is motivated not for political parties own sake, but because parties
are assumed to be essential for democratic political systems. Whether
the objective of IPA is defined as improvement of individual parties or
party systems, the ultimate goal is to make parties and party system
self-sustainable so that IPA becomes superfluous 36. At what point in
the development of a new democracy this stage is reached cannot be
specified. Even if political parties seem well-established and have every
indication of being institutionalized, both as individual parties and as a
party system, they may also become de-institutionalized. Venezuela
was seen to have a competitive party system in which two main parties
appeared to be well institutionalized (Myers, 1990). Nevertheless, to
36
This is not to argue that parties do not need international linkages. Transnational
federations, like Socialist International (SI), the International Democrat Union (IDU),
which organizes Conservative parties, and Liberal International (LI) are important
promoters of norms for parties in democracies, engage in the development of political skills,
diffusion of information and may also assist in protection parties against harassments by
political authorities. Transnational party federations in the European Union are particularly
important as they are linked to transnational party groups in the European Parliament.
57
the surprise of political scientists this party system collapsed
(Morgan, 2011) 37. In Mainwaring’s words: “The classic literature on
parties never contemplated the possibility that a highly
institutionalized system might collapse” (Mainwaring, 2012). Italy also
is an important reminder that even in well-established democracies,
the party system is not immune from collapse (Pasquino, 2006).
There are several examples, as we have seen that IPA donors have
withdrawn from countries in which they became involved. Exit was
chosen for a variety of reasons. IPA has in some cases been tied to the
donor country’s official development assistance (ODA), and when
official ODA priorities shift between countries, IPA has been
discontinued or greatly reduced. In other cases, funding has been
available for only a specified number of years and the projects have
therefore ceased at the end of the period. There are also examples
where the political environment has made it impossible to work with
political parties, as has happened in authoritarian states. Finally, IPA
projects have also ended because evaluation reports found them to be
inadequately organized or implemented.
The closest one comes to the choice of ‘exit’ because a state of
‘self-sustainability’ had been reached is in the former East-Central
European countries which have become integrated into the EU. In
these cases party assistance has been greatly scaled back. After an
initial period of fluid party system with weak institutionalization
(Lewis, 2000; Wyman, White, Miller, & Heywood, 1995) several
authors now have pointed out that political parties in the new
democracies in East and Central Europe do not differ substantially
from those in the established democracies in terms of ideological
orientations, strength of the respective party families (Bardi et al.,
2010; Hlousek & Kopecek, 2010) or in terms of organizational style
(Biezen, 2005)38. But the multidimensional nature of EU integration
has no counterpart in other new democracies. Other new democracies
are of course also partly involved in international networks and benefit
from ODA contributions in many fields, but the comprehensiveness of
EU-integration has no parallel in other regions. The attractiveness of
EU membership provided both ‘linkage’ and ‘leverage’ mechanisms in
37
According to (Lupu, 2014) between 1978 and 2007 a quarter of the established parties in
Latin America became uncompetitive for national executive office.
38
But see (Powell & Tucker, 2014; Sikk, 2005) for a more nuanced view of party system
stability.
58
EU’s democracy promotion (Lavenex & Schimmelfennig, 2011a). In
addition, the new EU states also have considerably better socioeconomic conditions than what is found in most non-EU new
democracies.
Few donor reports or academic studies find clear effects of IPA.
But is this finding the result of the methods used to investigate the
issue? Practically all evaluation reports call for a more systematic
collection of data in order to compare the state of the parties or party
system prior to the introduction of IPA and after the IPA has been
running for several years. But it is equally important to look at the
individual IPA actor in a wider perspective. In many countries, there
are several IPA donors. Although they most of the time work
independently of each other, there could be cumulative effects. Yet,
most studies of IPA’s, and this goes particularly for evaluation
reports, concern the contributions that a particular IPA institution has
made. It is not obvious that there will be positive impacts when several
actors are involved, but it is possible that there will be. More important
is that IPA is one of several forms of democracy assistance. Studies of
IPA should therefore consider possible interaction effects across
programs. Given the problem of data availability these types of
analysis call for case studies or studies with a few numbers of cases.
5.4 Key factors to consider
The review of the findings from evaluation reports and research on
IPA has identified a number of factors that may undermine a positive
outcome of IPA. Success or not is of course also contingent on the
standards against which IPA is measured and what the objectives of
the individual projects are.
On the one hand success of IPA projects depends partly on the
environmental factors, , and partly on the projects themselves. Some
of these factors, such as ownership, seem to be necessary for successful
IPA projects, others may be conducive for positive impacts, but
neither of the factors are necessary and sufficient alone. Thus, the
listing of the factors should be read as hypotheses for their potential
impact, not statements about actual findings.. The factors listed below
indicate some of the variables that both evaluation reports and the
research literature have identified as important, but their relative
significance is more problematic to assess.
59
Contextual factors
Political and institutional environments
Countries that lack freedom of information, where the authorities
harass opposition politicians and civil society activists, and where
there is a biased legal framework and political interference in the
application of the law, efforts to promote party and party system
institutionalization face greater challenges. One cannot expect that
support for (opposition) parties in such circumstance will yield
positive change in the short term. Resistance from political authorities
in such systems is likely to delay positive outcomes and therefore call
for long term engagements. In such cases support for CSOs may even
be more efficient than working directly with political parties.
But also the structure of political institutions may be more or less
favorable to IPA. Presidential systems, it is argued, are less favorable
to building strong party organizations than parliamentary systems
(Samuels & Shugart, 2010; Shugart, 1998). The combination of strong
presidencies without the possibility for re-election, is a particularly
bad combination which discourages political actors from adopting a
long-term perspective on organizational development, as in the case of
the Philippines (Hicken, 2009). Thus, because of the significance of
the institutional context van Cranenburgh argues that “… if the
systemic incentives toward personalism and one party rule inherent in
presidentialism are not addressed, these efforts (that is IPA) will not
have great impact” (Cranenburgh, 2011:454).
It is unlikely that IPA could have significant effects under such
institutional circumstances. In contrast, in parliamentary systems the
executive emanates from the legislature and depends on the legislature
for its functioning, which stimulates long-term cooperation among
MPs.
The electoral system also impacts on the need for party
organization, and is therefore likely to impact on IPA projects.
Proportional electoral system requires coordination, at least at the
regional/constituency level, but single member constituencies and
plurality elections do not. If in the latter case nomination procedures
are decentralized to the constituency level, are unclear or only
rudimentary implemented, the hurdles against party building is
considerable. It is of course, the legitimate right of individual
60
countries to choose whatever form of government (presidential, semipresidential, parliamentary) or whatever type of electoral system, it
prefers, but this should be taken into account when IPA projects are
designed and implemented. Such factors limit what it is reasonable to
expect IPA projects to achieve.
In addition to the formal institutions of governments, also other
institutions in the environment of parties, particularly mass media, and
electoral management bodies as well as the legal foundations for
political parties, constitute important constraints on how parties
organize and work. IPA projects should therefore be coordinated with
other democracy assistance projects because the prospects of a
predictable and level playing field constitute important incentives
party actors. Such environmental factors mean that IPA projects must
be ‘tailored made’ to the individual contexts in order to contribute to
the higher goal of democratic consolidation.
Socio-economic and cultural conditions
Creating and sustaining party organizations require lots of resources,
at the societal level as well as at the state level. Low level of education
tends to work against the building of strong formal organizations.
Low economic development and widespread poverty mean that there
are few economic resources that can be mobilized to sustain complex
organizations, while the combination of dispersed population
settlements and multiple culturally based cleavage lines (religious,
linguistically and ethnic) are barriers against effective communication
and mobilization, which parties depend on to function. To build
sustainable party organizations require mobilization of resources.
Deep conflicts along ethnic, religious and linguistic cleavage lines are
impediments against the building of national party organizations. IPA
projects alone cannot overcome such hurdles, but such factors need to
be incorporated into the calculation of what the expectations of IPA
projects should be. An unintended consequence of IPA in such
circumstances could be to create a dependency on foreign donors, as it
often alleged to be the case for civil society support. To alleviate such
challenges it is pertinent to ask if parties have strategies for mobilizing
resources? Should public subsidies for parties be introduced, and if so,
how?
61
Project factors
Ownership
IPA is a form of cooperation between donor(s) and recipient(s).
Cooperation requires commitments from both sides. The evaluation
reports and party analysis are quite consistent in pointing out that IPA
projects are not likely to succeed without the recipient parties
perceiving the IPA project as something worthy of pursuing. Yet, IPA
projects often find themselves in a paradoxical situation. The
motivation behind IPA is to contribute to change, but IPA projects
often have to work with political elites that benefit from the status
quo and may therefore not be interested in change. Even if the target
for an IPA project is not the elites in the central party organization,
but for example parties at the local level or the youth organization, the
implementation of projects must be accepted by the party leadership.
Initiation of IPA projects without the commitment of decisionmakers, particularly in the recipient party/parties therefore often run
into a barrier. It is therefore important that those actors who can be
considered as ‘veto players’ in the parties are actually committed to
working towards the objectives of the projects.
Institutional commitment
IPA programs embedded in the organization, both among the donor
and the recipient party/parties, is more likely to succeed than if the
IPA project primarily relies on individual actors and not on
organizations. It is of great value to any projects that there are
individuals who are strongly committed. But reliance solely on
individual persons may make projects more vulnerable when there is
turnover among persons in the party organizations. The involvement
of several actors may provide better opportunities for the sharing of
information and provide opportunity for the mobilization of different
types of skills. Thus, cooperating parties should have routines for
dissemination internally in the parties of what the IPA project is about
and involve several actors.
62
Comprehensive approach vs. short term projects
It seems to be the unanimous opinion among donors that party
assistance takes a long time to have any effect. Individual projects
should therefore be embedded in more comprehensive programs and
short term projects or single event projects should consider how
follow-ups can be organized.
Pending on what the objectives are, it is not possible to identify if
small projects impact on overarching goals, such as party system
stabilization, but smaller projects can contribute to clearly specified
goals. It is also possible that the cumulative benefits of smaller
projects can be positive for overall goals if they are all pulling in the
same direction. This, however, requires coordination and it is a
challenge in IPA projects to know how significant IPA funds are
compared to other funds. An additional challenge for small projects is
that the administrative costs of running the project may take a large
share of the available funds.
Coordination
IPA is one of several types of efforts to assist democratization
processes, which in turn is part of a more general developmental
agenda. The institutionalization of individual political parties and of
the party system is strongly impacted by institutions and processes in
the political system, such as the organization of elections, the legal
regulations of parties and elections, the strength of the parliament and
the judicial system. It could therefore be argued that IPA projects
whose objectives it is to contribute to the overarching goals of
democratic stability or party system institutionalization, should be
linked to, or coordinated with development assistance projects that
target factors in the environment relevant for parties.
Monitoring, evaluation and adjustments
All development efforts are in some ways a form of trial-and-error.
This may particularly be the case in models based on party-to-party
projects which involve actors that have IPA as a side activity. It is also
understandable that in the early phases of an IPA donor’s existence,
there is a search for methods that may work. But unless there is a
63
system for monitoring and evaluation of projects there is no basis for
making necessary adjustments. In such cases donor institutions and
recipient parties would lack necessary information to adjust the
programs. Therefore, IPA should start with the collection of
benchmark data prior to the introduction of the programs. Such
information should be collected continuously throughout the project
period. But such information requires a common understanding
between donors and recipients; hence the importance of trust among
the participating organizations and individuals. While projects may
end after a certain period, even when successful, it is important that an
institution, for example the funding agency or, has in place a system
for continuous monitoring for some time after the end of the project.
This is necessary in order to assess the sustainability of what IPA
projects have achieved.
Reflections
All of the factors above can be considered variables that impact on
whether IPA works or not, but apart from the issue of ownership, it is
unclear how each of these variables affect the success of IPA projects.
This is because the objectives of IPA are at different levels – from the
meso-level, improving particular parties, improving all parties, to the
macro-level improving the party systems, and improving democratic
consolidation. There are fundamental gaps in our knowledge how the
seven factors are connected but also generally a shortage of
information. A further finding is that many of the objectives of IPA
are strongly impacted by the institutional and socio-economic
environments, which call for greater coordination between IPA and
other democracy assistance efforts and also for the need to be flexible
as to the targets of IPA projects.
64
Appendix A: Findings from research
literature on IPA
Tom Carother’s volume “Confronting the Weakest Link. Aiding
Political Parties in New Democracies” (Carothers, 2006) is the most
comprehensive study of IPA. The essence of his study has been
summarized in the formulation that “It rarely has transformative
impact, despite the hopes and sometimes belief of its providers”
(Carothers, 2006:160). The reasons for this is that “the main
determinants of the make-up and methods of parties in the recipient
countries are an array of underlying economic, political, social, and
cultural conditions that are largely beyond its ambit” (Carothers,
2006: 214-215). Transformative effects are, for Carothers when
recipient parties “substantially ameliorate the central organizational or
functional characteristics that the aid providers seek to
address”(Carothers, 2006: 163). An example of this would be if
leadership fixated party without a grass-root organization was
transformed into a membership based organization with internal
democracy. Carothers find that there are some examples of change
due to IPA, most notably in campaign methods There are also cases of
small organizational changes, as when a party that has not held a
national convention actually do so. “The hard question is whether the
small positive steps will deepen over time and cumulate in major
organizational changes” (Carothers, 2006: 187). There are more
examples of projects that have failed to achieve their objectives.
Attempts to induce parties to change are met with opposition from
party leaders who fear they will lose control of the parties. In other
cases the environment, both the institutional (for example the role of
the monarch in a polity) or economic (such as wide spread poverty)
works against efforts to create well- structured parties.
Since Carother’s book was published, there have been several
studies, primarily of individual countries or comparisons of party
developments in some countries. Have these studies confirmed
Carother’s finding or do they arrive at different conclusions?
Case-studies and comparative studies of IPA
The edited volume by Burnell and Gerritt (2010) is the most recent
compilation of research on IPA. The articles cover assistance
65
programs in the Balkans, Central Asia, Latin-America, countries in the
Middle East and in Sub-Saharan Africa. Moreover, the IPA programs
in these countries have been provided by a variety of donor
organizations and for different time periods. The programs also vary
according to the objectives, although strengthening the party system
and individual party organizations figure most prominently. Most of
these articles conclude that IPA does not work, or that IPA has
unintended effects, while others argue that there are some examples of
positive outcomes. Although this is the most recent compilation of
IPA studies, political developments since they were written may
indicate whether or not their conclusions are still valid.
The title of Bader’s (2010) article – “Party politics in Georgia and
Ukraine and the failure of Western assistance” - sums up his view. He
argues that in these two countries party assistance has failed to
stabilize the party system and also failed in promoting party
institutionalization. He characterizes the parties as ‘notoriously
undemocratic concerning their inner functioning’ (Bader, 2010: 1097).
Parties included in the assistance programs have not been able to
survive. The inclusion of parties in assistance programs ‘have been
driven by misguided perceptions of those parties’ (Bader, 2010: 1098)
Developments in these countries since Bader’s analysis was conducted
have confirmed the fluent nature of the party system, but it also shows
that conclusions based on IPA prior to 2012 may not necessarily be
applicable to developments after 2012. In the case of Georgia, the
changes have been dramatic. The decision by Georgia’s richest
business man, Bidzina Ivanishvili, to form his own party, Georgia
Dream, shortly before the parliamentary election in 2012 overturned
the political landscape. The outcome of the parliamentary election that
year (and the presidential election the following year) was a complete
surprise (Waal, 2012). However, given the short time that has elapsed
since the most recent elections, it is impossible to estimate whether
the new party system will prove more stable or the parties be more
internally democratic, than in the pre-2012 era. Nevertheless, there is
at least one example – the Republican Party - that appears to be both
more internally democratic and institutionalized. The reason why
Bader is so critical of assistance program is that the recipient parties –
that is the party leaders – are not interested in developing a democratic
organization. Many of the parties that received assistance were not
likely to survive, but the donor’s seemed unconcerned with selecting
66
parties that were likely to survive. (This would in any case be hard to
predict).
A somewhat similar negative conclusion is also offered by Hulsey
and Nenadovic, who have studied the international community’s
efforts to strengthen multi-ethnic and non-nationalist parties in
Bosnia-Herzegovina. (Hulsey, 2010). This type of parties is assumed
to be better suited for national integration than the ethnic and
nationalist parties. Yet, there is little to indicate that this type of
support has had much success; quite the opposite. The reason for this
Hulsey argues, is that the electoral system encourages political actors
to mobilize as much support as possible within ‘their own’ ethnic
group. Nenadovic argues that the international community has failed
to transpose its democratic norms to the local elites: “the local
political elites seem to have had more success in ‘socializing’ the
international administration officials to ‘the local ways of doing
things’ than the latter have had in making the local elites more
democratic and accountable in their work” (Nenadović, 2010:1168).
Spoerri (2010), on the other hand, finds a clear effect of party
assistance – but not one that is desirable! She examines the IPA
programs in Serbia in the post-Milosovic period and base her analysis
of 150 interviews with representatives from the donor community and
the recipient parties, as well as documents made available by the donor
organizations. The argument in the article is that IPA in Serbia has not
been directed toward stabilizing the party system or seeking to
improve the internal democracy in parties, but rather to achieve a
particular political outcome. In the post-Milosovic period the main
goal for the donor community was to prevent parties associated with
the old regime of winning elections; regardless of any changes that
might have taken place in those parties. She argues that donors viewed
the parties through the same lenses (democratic vs. anti-democratic)
as during the Milosovic reign and failed to adjust their strategies to a
changing political landscape. Thus, in her view external actors meddled
in domestic processes, exacerbated political polarization and created a
culture of dependency. Although it is impossible to demonstrate that
it was donor contributions to particular parties that determined the
outcome of the elections, the objectives of the donors were precisely
that. Whether or not the recipient parties contributed to democratic
consolidation, a stable party system or internally democratic parties
were beyond the donors’ concern.
67
Bolleyer and Storm are similarly hesitant about the efforts to
promote parties, in their case in Morocco (Bolleyer & Storm, 2010).
They studied how the two American party assistance providers, NDI
and IRI, and the multilateral UNDP, had worked during a decadelong period. They identified nine potential targets of individual party
change, and 11 potential party system or inter-party relations targets.
The former included such items as institutional characteristics of the
parties, party offices, training of party officers, regulations of intra
party relations and the latter party access to state funding, electoral
rules, policy orientation and communications between parties and
citizens. The types of change that IPA could potentially lead to were
grouped into three main categories: a) change in formal-legal incentive
structures, b) the functioning of institutions, and c) societal relations.
Thus, donor efforts to promote party developments covered an
extensive range of topics. However, in spite of the efforts by the three
donor institutions (and in addition three German party foundations
are also involved) “..there are not many positive implications of these
organizations spreading themselves so thinly”(Bolleyer & Storm,
2010: 1213). The spread of efforts on so many activities is one of the
reasons why so little progress has been made. A more concentrated
program would be better because of the need to overcome two basic
impediments to party building: the incumbent party elites and the
executive monarchy.
Moroccan party elites are not really interested in developing parties
towards some form of a European mass party model, they argue,
because it is not election results that determine the elites’ access to
power. What counts in Morocco is the executive power controlled by
the monarch. Elections are held, but the monarch decides who gets
access to what. Morocco has made some changes in recent years away
from a traditional, authoritarian system, but the monarchy is
interested in stability, not in change. The authors dismiss these
changes as ‘cosmetic’ (Bolleyer & Storm, 2010: 1216). Party elites are
very motivated to take part in the donor programs, “…but largely for
the wrong reasons”(Bolleyer & Storm, 2010: 1215) . The incumbent
elites are strengthened by the influx of resources. This situation has
left the Moroccan electorate greatly dissatisfied with political parties.
The authors therefore argue for a concentration of party promotion
on developing links between parties and civil society. In the long-run
this may generate grass-root demands for genuine democratization –
but precisely because of this, donor agents may be hesitant in getting
68
involved, particularly in a state considered to be a stable ally of
Western governments.
In the same volume, Kristian Weissenbach analyzes the Friedrich
Ebert Stiftung (FES) 39 activities in Kenya and South Africa
(Weissenbach, 2010). The objectives of FES changed over time. In the
period prior to democratization, FES worked to identify and assist
pro-democracy actors, whereas after the democratic transition (in
Kenya 1992 and in South Africa 1994), the objective was to strengthen
political parties. Weissenbach identifies seven criteria of party
institutionalization: 1) level of organization, 2) internal party
democracy, 3) programmatic, 4) autonomy, 5) roots in society, 6)
coherence, 7) regional and international integration ((Weissenbach,
2010: 1230). To measure these criteria she applies Basedau and Stroh’s
operationalization of the framework outlined by Randall and Svåsand
(Basedau & Stroh, 2008; Randall & Svåsand, 2002). The purpose is to
study how FES’ approach to party cooperation varies along these
seven criteria. The research is based on extensive field work in the two
countries, particularly interviews of the resident representatives of the
German party foundations and with representatives of the recipient
parties.
Prior to the end of the Apartheid regime in South-Africa, FES had
supported ANC (African National Congress) for decades. This
cooperation continued after 1994 and FES saw the ANC as an
ideological sister party. FES was in particular, instrumental with
regard to the 7 criteria: improving ANC’s acceptance in the Western
hemisphere. While ANC is clearly institutionalized in several of the
criteria, it does not score as well in terms of internal democracy. But it
is not so clear what FES can do to change ANC in this respect.
And, while ANC is clearly institutionalized, many donors and
analysts of South African democracy are more concerned with the
structure of the party system. South Africa is a prime example of a
dominant party system (Southall, 2001). 40
39
FES is the research foundation of the German Social Democratic party.
It should not be assumed that a dominant party system is by definition negative for
democracy. Sweden, Norway, Japan and Italy have all been through periods of as a dominant
party system, but few would argue that these countries were less democratic than countries
with alternations of parties in government. There were, however, counterwailing forces
(extensive local democracy, corporatism (Sweden and Norway) that limited the impact of
national government. In South Africa elements of federalism may have a similar restraining
effect of the national government.
40
69
In the case of Kenya FES worked with KANU (Kenyan Africa
National Unity) in the 1980’s. The cooperation was extensive. FES
provided campaign material, built a party headquarters and supplied
infrastructure support. Nevertheless, FES was unable to change
KANU into a more democratic organization and was not able to find
proper cooperating partners within KANU. As a consequence, FES
withdrew from the cooperation and from party assistance in general in
Kenya. It is only in the recent years that the organization again
initiated support for several of the new parties. In contrast to the
South African case it has not been possible for FES to identify a
similar ideological sister party in Kenya. Hence, a multiparty approach
has been seen as more relevant. FES has worked with five parties.
However, all of them score very low on the indicators of
institutionalization. It is not possible to conclude that the lack of
institutionalization is because of the short time that has elapsed since
IPA was re-started, or because of the ‘personalistic’ nature of the
Kenyan parties.
Also Rakner and Svåsand find limited effects of NIMD’s activities
in Malawi and in Zambia (Rakner & Svåsand, 2010). As outlined
earlier in the report, NIMD has three main objectives: improving
inter-party dialogue, improving party institutionalization, and
improving party-civil society relationships. In the case of Malawi and
Zambia, the first objective has been achieved in the sense that parties
with representation in parliament have agreed to participate in a
multiparty dialogue forum. The long-term objective of such fora is to
improve the relationships between competing parties. As such the fora
have succeeded by bringing together parties that otherwise do not
interact with each other. Nevertheless, inter-party dialogue has not
lead to the stabilization of the party system, as in both countries party
fragmentation has continued and party elites defect from one party to
the other or form their own party.
Participation in inter party fora is a requirement for parties to
receive direct support to develop their party organizations. In this
respect there is less clear effects of IPA. Parties in both Malawi and
Zambia are heavily dominated by the central party leadership and are
still lacking in establishing routinized procedures for basic party
functions, like nomination of candidates for parliamentary offices and
for arranging regular meetings of the units in the party organization.
In several cases Zambian and Malawian parties also failed to comply
with the reporting requirements for receiving direct party support,
70
which indicate poor administrative routines. In the end, NIMD
withdrew from Zambia after the failures of the inter-party dialogue
forum to function as well. (In Malawi cooperation continued until
2013, when Dutch development aid no longer targeted Malawi). As in
the case of Morocco, the impediments against party building are partly
the party elites themselves, partly the incentive structure in the
political system. Party elites benefit from the existing nature of the
parties, hence have no interest in developing a large organization
which allows grass-roots to exercise control over the party leadership.
Moreover, in both countries the presidency is by far the most
important political office and in both cases the president is elected by
a plurality of the votes in a national election. It is simple to register
new political parties. Hence, ambitious politicians with their eyes
focused on winning the presidency will form their own party if they
are not nominated as presidential candidate in the party they belong
to, developments which have continued until this day. Party
fragmentation continues, with Malawi having more than 50 parties in
2014. The two last presidents have both come from parties caused by
fragmentation of existing parties. New – and some of the old – parties
suffer from incomplete organizational procedures. Hence, candidate
nominations for parliament have been chaotic in several parties
(Svåsand, 2014). In the Malawian case, only the old party of the one
party state, Malawi Congress Party, has a consolidated organization,
although still heavily leadership dominated.
Research on democratization where IPA is one of several factors
Glenn analyzed how the broad popular movements that were central
in the transition process in Czechoslovakia were replaced by regular
political parties. The ‘party system’ in the first election was dominated
by broad popular movements, rather than by party organizations
Competing actors inside the movements favored the development of
separate parties and IPA had a strong impact on the structure of the
emerging party system. (Glenn, 2000). Glenn found that the National
Endowment for Democracy (NED) was the dominant foreign donor
during the transition in 1990 and accounted for almost 40% of the
total sums spent on democracy promotion. By far, most of the money
was spent in what was to become the Czech republic. NDI in
particular responded rapidly to requests from the newly elected
president, Vaclav Havel. Glenn concludes that assistance during such
71
transition years “…can have disproportionally large consequences
because the institutional arrangements created in this period can
structure the medium-term allocation of resources among political
contenders”(Glenn, 2000:175). After 1990 there was a sharp decline of
NED’s funding for Central and Eastern Europe and international
finance for parties was partly replaced by public subsidies for parties.
Irvine (2013) study how women become better represented
politically in Croatia and in Serbia. She credits international assistance,
particularly American, with having had a major role in this change. US
initiatives helped oust authoritarian leaders in Croatia and in Serbia
because of its ‘investment’ in ‘cooperation among opposition parties
and between political parties and civil society’ (Irvine, 2013:248),
including supporting women’s groups with leadership training and
organizational capacity. In Serbia alone the Americans spent 50
million USD in 2000 for democracy promotion in general, most of
which went to an alliance of opposition movement (Irvine, 2013:249).
However, she argues that the parties themselves were not very
forthcoming on this issue. It was mainly due to strong mobilization
among women NGOs that increased female representation was
possible. Parties were more of an obstacle than a driving force.
72
Appendix B: IPA donors and findings
from evaluations
Below is an overview of a select number of IPA donors, their models
for IPA support and findings from some evaluation reports. The
selection of evaluation reports do not claim to cover all donors, or to
be a representative sample. IPA is organized in different ways and
implemented in different ways.
The selection is heavily influenced by the fact that IPA donors
have different practices concerning the publication of evaluation
reports, as well as by my own field of competence. NIMD (The
Netherlands Institute for Multiparty Democracy) is the most
transparent IPA donor organization. They have a long series of
evaluation reports, both covering NIMD as an organization and
evaluation of NIMD’s country programs (www.nimd.org). Other
IPA donors have less systematic practices, but nevertheless have some
evaluation reports the British Westminster Foundation for Democracy
(WFD) and the two American institutions, National Democratic
Institute (NDI) and International Republican Institute (IRI). There
are also some reports for the more recent IPA institutions in
Denmark and Finland. Swedish IPA programs have been evaluated
three times (SADEV, 2009; Uggla et al., 2000; Öhman et al., 2004)
and is currently again evaluated. The German Stiftungen are the least
transparent. No evaluation reports similar to that of other IPA donors
are available.
The evaluation reports summarized below represent IPA donors
that vary along several dimensions: organization, objectives, funding,
countries where they work (or have worked), and for how long they have
worked. Moreover, the evaluations have also been done at quite
different time points. Some have been executed a few years after the
establishment of an IPA donors, others evaluated projects that been
running for several years. But taken together there are several findings
that cut across the different models and therefore can be useful in
identifying the problems involved in IPA and the factors that may
contribute to IPAs efforts vs. those that obstruct them.
I start with summarizing the findings to previous evaluation
reports of the Swedish PAOs. Are the results for the Swedish PAO
similar to, or different from projects organized by other donors that
73
also use the party-to-party principle, such as the Danish Institute for
Parties and Democracy (DIPD), or the Norwegian Center for
Democracy Support, which existed between 2005 and 2009?
Compared with the this approach to party-to-party assistance,
what are the conclusions from evaluations of donors that primarily is
based on a multiparty approach, such as the two American
institutions, the British, and the Dutch IPA donor.
Sweden: support through party-affiliated organizations
Swedish IPA started in 1995 as a temporary measure which was made
permanent in 2001. IPA is an element in Swedish democracy
assistance. The overarching aim is: “Democratic development and
increased
respect
for
human
rights
in
developing
countries”(Regjeringen, 2010:1). As part of this goal the aim of IPA in
the initial phase was to “contribute to the development of a wellfunctioning and pluralistic party system and democratic societies in
developing countries and in Central and Eastern Europe41” (cited in
(Uggla et al., 2000:3-4).
IPA in Sweden has been based on ‘the sister party model’. Parties
with representation in the Riksdag have established foundations
handling IPA, so-called PAO (partianknytna organisationer). The
financing of PAOs has been based on the respective parties’ share of
the seats in the Riksdag, plus a basic grant irrespective of the party
size. Thus, it is not the priorities of Swedish development policies, or
the needs assessments of recipients that determine the funding pattern
across PAOs, but the preferences of the Swedish electorate. In 2012,
party assistance was more clearly differentiated into a) support for
sister parties and b) support for multiparty projects. The former is
based on program proposals for a three year period and funding for
these are based on the PAO-parties representation in the Riksdag. The
latter objective is supported based on competitive applications to
SIDA (SIPU, 2011).
Swedish IPA has been evaluated several times. The evaluations have
covered the initial period 1995-1999 (Uggla et.al, 2000), the period
41
This, and other quotations from the report, is my own translation
74
2001-2004 (Öhman et al., 2004) and the years 2007-2008 ((SADEV,
2009).
Evaluation of Swedish support in 2000
The main analysis of the PAO report concerns the relationship
between input (regulations and structure of the support), outputs (the
specific activities funded) and outcome (whether or not the projects
contributed to the stated objectives).
As for the first, the PAO were autonomous in choosing whom to
support and how. Consequently, a wide variety of projects, and
reporting of such, existed between the PAOs. The report notes that
funding had nothing to do with how projects were chosen or
executed, as the funding was tied to the mother parties’ electoral
performance in Sweden. Most of the outputs concern support for
policy and ideological developments, candidate recruitment and the
strengthening of party organizations. Linkages between parties and
civil society ranked low in priority. Recipient parties appreciated the
cooperative nature of the support and local ownership was high.
According to the evaluation, the projects had contributed to
strengthening the recipient parties as organizations 42, but there was
little concern among the participants for the second objective of IPA,
improving the party system. The evaluators indicated that given the
sister party model it would be difficult to expect the IPA to contribute
to strengthening the party system and the fact that a large share of the
support went to small parties, the opposite – fragmentation of the
party system – might be an unintended effect. The autonomous nature
of the PAO meant that selection of projects and parties depended to a
great extent on established contacts. Hence, support for parties was
allocated to countries without an assessment of where the need for
support was strongest.
As the parties’ enjoyed significant autonomy in deciding on
projects, and given the sister party model, the report found 45% of the
total funding was spent on parties receiving less than 10% of the votes
in the previous elections. Thus, the funding model, given several small
parties in Sweden, could mean that IPA contributed to fragmentation
42
However, there are no examples or data about what it is that has changed as a result of the
IPA.
75
in the recipient countries (Uggla et al., 2000:23). Geographically, for
the whole period under evaluation 2/3 of IPA’s funding was allocated
to projects in Eastern and Central Europe (with the Baltic states
topping the list of recipient countries), with 1/3 going to developing
countries. This geographical bias was almost 4/5 in the first years.
In addition, the evaluators pointed out some administrative
challenges. There was little expertise in this field in the PAO which
were more “..characterized by party political competence than
development competence” (p. 21). There were also discrepancies
between the budgeting for projects and the actual implementation of
projects. Administration consumed almost 25% of the funds.
Evaluation of Swedish support in 2004
In the second report (Öhman et al., 2004) the evaluation team was
particularly concerned with how party assistance was administered and
with the linkages between the individual PAO projects and the overall
goal of the program “:..contributing to the development of a
functioning party system 43”. While arguing in favor of allowing PAO
autonomy in choosing project partners and projects, the committee
also recommended more coordination. Autonomy was important to
avoid that party support became too closely connected to official
Swedish development policy. On the other hand, the way party
support was conducted did not ensure a linkage with the overall aim of
strengthening the party system. Increased communication and
information sharing between PAO, SIDA and other relevant
institutions were proposed as ways to encourage greater cooperation
and also to increase the linkage between party support and other
forms of democracy assistance. There was also a need to increase the
budget allocated to cross-party activities, without reducing the
individual PAO share 44. Concerns were also expressed about the
spread of PAO projects across 50 countries in 2004 (Öhman et al.,
2004:29).
As was the case with evaluation reports for other countries (see
below) the evaluation team also called for a development of more
systematic assessments of the impacts of the projects.
43
44
«partivasende» in Swedish, in contrast to the 2002 term «partisystem».
At the time of the report the cross-party share was 8%.
76
Evaluation of Swedish support in 2009, including three cases
The third report (SADEV, 2009) covered the 2007-2008 period. It
consists of an overall study of the 157 projects conducted in 39
countries in this period, but also of three case studies: Uganda,
Colombia and Ukraine. Some of the concerns expressed in the earlier
reports are repeated in this one: while most projects are conducted
according to the expectations, there are weak linkages between the
projects, which are for the most part targeting sister parties, and the
overall aim of strengthening the party system. And, as in the earlier
reports, this one also noted that the level of information and
coordination between various actors could have been better.
Moreover, the report argues that the number of projects and countries
was too high compared to the available funds. The three case studies
reveal contrasting results.
Colombia: The report concluded that at the project level the party
support had been successful (to the extent that the data allows for a
conclusion) but “..it is not possible to trace any impact at the party
system level” (Peck, Rudqvist, & Ramos, 2009:6). Colombia
illustrated also, according to the report, how individual projects could
be successful at the output level without contributing higher-level
goals. The report notes, for example, the lack of linkages between the
overall Swedish development policy and the projects organized by the
PAO. Another example of problematic assistance is the fact that two
Swedish PAO supported different factions of the same party. Thus,
this may have exacerbated intra-party divisions, while the party that
was best positioned to challenge the incumbent party in Colombian
politics, however, did not receive any support at all. There were also
several problems related to project management. Most projects were
small (financially) and were lacking in a long-term perspective, weakly
institutionalized because they were heavily linked to individual actors,
and with weak information sharing with the Swedish embassy. Weak
information sharing contributed to the lack of coordination.
The evaluation report found some projects to be functioning quite
well according to the objectives, while others had significant failures.
The support for the Columbian Communist party (PC), at the time a
part of the new united leftist group, was deemed highly successful as it
had contributed to maintaining and strengthening an institutionalized
party that challenged the concentration of power in the presidential
office. At the other extreme is the Green party’s involvement which
77
was found to be lacking in focus and with weak connection to the
overall objectives of party assistance. Many of the projects suffered
from inadequate planning, execution, and reporting. (These problems
were also present to some extent in other more successful support
arrangements).
The report on the PAO in Colombia identified a number of
challenges and therefore variations in how successful the projects
were. A major problem that runs across (almost) all projects is the
weak or non-existent linkage between the projects and the overall
objective of strengthening the party system. In many cases project
formulations were vague or changing over time and lack in attempts to
document how the projects had contributed to the stated goals. Lack
of information and knowledge of the political context was also
common and only in a few cases did projects have a close follow-up
from the Swedish PAO. The report questions to what extent the
support was sustainable. Few cases existed where follow-up projects
had been planned. To some extent, the lack of outcome were the result
of changing political circumstances which were beyond the projects
control, but the major weakness was in project design and
implementation.
Uganda: Another country report covers the PAO’s involvement in
Uganda, 2007-2008 (Kjellström, Makara, & Sjöberg, 2009). This
involvement followed a completely different model. Four of the
PAO’s were engaged in a joint project working with five Ugandan
opposition parties. In addition, one of the PAO’s (KIC) worked with
the American organization NDI in a cross-party project targeting
young politicians. The former project was deemed partly successful
because it helped to promote communication and trust between
political competitors and it was largely successful due to its
implementation via a local NGO. Communication between this
project and the Swedish embassy was also good. The creation of the
IPC (Inter Party Committee) was seen as an instrument to unite the
opposition parties in front of the general elections in 2011. The
ambition was to agree on a joint presidential candidate and coordinate
the nomination of MPs. A drawback however, was that the projects
did not have any plans for the post-2011 years and the evaluation team
therefore questioned the sustainability of the arrangement. Later
developments proved the evaluation team right. The opposition
parties were not able to unite behind a common presidential candidate
and the coordination of MP nomination was equally unsuccessful.
78
However, inter-party cooperation continued; not because the IPC
survived, but because of the creation of IPOD (Inter party
organization for democracy); an outfit under the umbrella of the
Dutch NIMD, but now enjoying multinational support.
Ukraine: The findings of the Ukraine-report point to the paradox of
the PAO model in finding relevant partners and the structure and
challenges of the Ukrainian party system. The Swedish PAOs
identified parties (and in one case an NGO) that were seen as
ideological sister parties. However, these parties were fairly marginal
in the Ukrainian party system and the future of some of them
uncertain. On the other hand the Ukrainian party system was strongly
in need of consolidation, both as a system and for each of the parties
as organizations. The evaluation report noted that although
participants in the activities appreciated the support programs it was
not possible for some of the activities to see how participation had
improved the party organizations. There appeared in some cases, to be
no links between the programs and what went on in the parties. No
project evaluations had been carried out and follow-up plans did not
exist. The report notes that reliance on information and knowledge as
a major stimulus for change ignored that established elites may benefit
from things being the way they were (Peck, Kjellström, Kuzio, &
Sjöberg, 2009: 7-8).
In general, because of the Swedish model, based on PAOs seeking
to support relevant cooperating parties, the main problems in the
structure of the Ukrainian party system – fragmentation, lack of
internal democracy, corruption and defections by party
representatives from one party to another – were not addressed. (As
noted above, the Uggla-team did warn against the party-to-party focus
already in its 2000-report.)
Concluding reflections: The three country reports reveal varied
degree of success, within countries as well as between them. Thus, it
would be incorrect to conclude that the PAO model is always
successful, nor is always un-successful. Even within one country, some
PAO activities were found to be more successful than others. A
general problem identified by the country reports is the lack of
linkages between what the PAOs do and the overall objective of the
IPA programs: to improve the functioning of the party system. This
lack of linkage seems to originate in the strong preference in the
program for PAOs’ own identification of partners and specific
projects. However, the projects themselves often seemed to lack clear
79
ideas of what their objectives should be and how to ‘measure’ progress
towards those objectives. Several problems were also identified in the
implementation and reports of the projects, partly resulting from lack
of knowledge, insufficient communication with the stakeholders and
the absence of continuous representation in the countries. The
Uganda experience was, on the whole, more promising than the two
other countries. Here, the PAOs worked together and the recipient
parties did also cooperate 45.
In all three cases, success - or lack of such - should be seen against
the back drop of the local political context and the scale of the
Swedish support. The legal and institutional environments are factors
beyond the influence of the PAO model. Swedish support is one of
several types of IPA and it is a big unknown how much IPA in general
account for the parties’ total finances. Moreover, the evaluation
reports cover a short time span in each country and as pointed out, in
many cases it is not possible to conclude that projects have had an
impact or not. Outputs, such as activities can be documented, but the
linkage between output and effectiveness cannot be assessed.
It should be noted that several of the criticisms in the 2009
SADEV reports have since been addressed by Sida. There is more
emphasis on cross-party initiatives and the quality of the applications
has improved. 46 Project objectives are more clearly specified, with
relevant indicators. At the same time, the main model for organizing
IPA is on party-to-party assistance and less on party system
developments. The variations found in how successful projects have
45
In Uganda, Sweden, together with several other donors (Austria, Denmark, Ireland, the
Netherlands, Norway, the UK and the Delegation of the European Union in Uganda) has
joined in basket funds. Nevertheless, there is little that indicates the program has been
successful in the sense that Uganda’s political system has become more competitive. The
opposition parties continue to be fragmented. There were more political parties contesting
the parliamentary elections in 2011, but as in 2006 most of the parties were not able to
penetrate the territory with candidates. The 2011 election reinforced the results from 2006
.Rather than becoming more competitive, NRM’s dominance was enhanced further. Were
these results to be repeated in the next election in 2016, Uganda would be clearly a dominant
party system. Uganda scored worse on Freehom House political rights (6) in 2013 than in
earlier years (5). The context for a stable party system is unfavourable. Uganda has several
ethnic and religious divisions that intersect with the party system. The Constitution
recognizes 65 indigenous ethnic groups and is one of the worlds’ most ethnically
fractionalized states (Green 2010).The social and economic conditions are poor: Uganda was
ranked as 163 out of 187 on the UNDP Human Development Index.
46
Sida, 2014a, 2014b.
80
been in terms of strengthening party developments is partly resulting
from the design and the implementation of the individual projects, but
local contextual developments can also undermine the best of
intentions. The environments of parties in new democracies, such as
the legal regulation of parties, the quality of elections and the
structure of political institutions are largely outside of the PAO’s
activities but nevertheless influence their success.
As the following review of other IPA models reveal, the Swedish
experience, and other IPA donor working with the party-to-party
approach, is not unique. Successful and less successful examples of
IPA can be found, regardless of organizational models as can be seen
from the evaluations of the American, British and Dutch donors.
Denmark: The Danish Institute for Parties and
Democracy
Although Denmark is a major development donor, Denmark has not
until recently become involved in IPA. The DIDP (Danish Institute
for Parties and Democracy) was established in 2010 and is funded by a
three year grant (DK 75 mill.) by the Danish state budget. The budget
is split 40% for bilateral projects for the parties and 40% for
multiparty projects. The former is run by the individual parties, while
DIPD runs the multiparty projects (DIPD, 2013). Of the funds for
bilateral projects, one third is shared equally between the parties
represented in the Folketing 47 and two thirds split according to the
parties’ share of the seats. As we have seen in the reviews of Swedish
IPA there have been calls for closer links to other elements of Swedish
development policy. In the Danish case there are 14 countries where
DIPD is active, Bhutan, Bolivia, Egypt, Ghana, Honduras, Kenya,
Malawi, Myanmar, Nepal, Palestine, Swaziland, Tanzania, Zambia, and
Zimbabwe, but minimal overlap with countries receiving most of
Danish ODA. But it is only three countries, Bolivia, Tanzania and
Zambia, that are also among the 14 most prioritized countries for
ODA(DIPD, 2013) (p. 15)48.
47
One of the eight parties did not apply for any IPA funds.
The others being Afghanistan, Bangladesh, Burkina Faso, Ghana, Kenya, Mozambique,
Nepal, Somalia, South Sudan, Uganda and Vietnam.
48
81
The division of the budget between multiparty and party-to-party
projects is a consequence of the purpose of the institutions: “The
vision of the Danish Institute for Parties and Democracy is to
contribute to the development of well-functioning political parties and
multiparty systems in a democratic culture, in support of the
aspirations for freedom and human development of citizens in
developing countries”(DIPD, 2013).
Evaluation of Danish support in 2013
DIPD was reviewed in 2013 (MFA, 2013). In seven countries
(Tanzania, Malawi, Myanmar, Bhutan, Nepal, Egypt, and Zimbabwe)
DIPD runs multiparty projects, in the two last countries in
partnership with NIMD. The party-to-party projects are only in a few
cases in the same countries where DIPD runs multiparty projects and
where they are in the same countries there seems to be little
interaction between the party-to-party projects and the multiparty
project. The individual parties run projects in a range of countries:
“The Danish Liberal Party is working in Kenya with the Centre for
Multiparty Democracy (CMD) 49, and in Zambia with the small
opposition party, the National Restoration Party (NAREP), in
Tanzania, the Conservatives and the Social Liberal Party have
partnered with the two largest opposition parties, CHADEMA and
CUF respectively, the Social Democrats have developed party-toparty programmes in Ghana, with the National Democratic Congress
(NDC), in Swaziland with Swazi Democratic Party (SWADEPA), and
in Egypt, with the Egyptian Social Democratic Party (ESDP). In
Honduras, the Red-Green Alliance is training party members through
its programme with the National Popular Resistance Front (FNRP).
In Bolivia, the Socialist People’s Party entered into partnership with a
hybrid political party, Movimento al Socialismo (MAS), to work on
organisational development and youth participation. The Liberal
Alliance has also been collaborating with the small and new opposition
party, Verdad y Democracia Social (VERDES), as part of its appraisal
to identify a party-to-party partner in the country”(DIPD, 2013:45).
The geographical spread on so many countries with a rather small
budget caused the evaluation report to advice for more concentration
in the future.
49
An institution created by the NIMD Kenya program.
82
Norway: The Norwegian Center for Democracy
Also Norway has experimented with IPA. In 2002 the Norwegian
Center for Democracy (NDS) was created as an institution to assist
new democracies and particularly the development of political parties,
roughly similar to the WFD model. However, it never succeeded in
establishing good organizational routines and after critical newspaper
reports of its activities, the Ministry of Development dissolved the
center in 2009. Party assistance has since been brought under
NORAD (Norwegian Agency for Development Cooperation). The
decision to close down the center is rather unique, compared to other
state sponsored institutions in general and to other IPA donor
organizations and precisely therefore it is relevant for this paper. So,
what went wrong?
Evaluation of Norwegian support in 2010 50
The evaluation report (NORAD, 2010) argued that Norwegian
political parties demonstrated lack of a strategy of partnerships on the
part of NDS. Most of the parties were interested in bi-lateral projects,
few in cross-party activities intended to strengthen the party system.
Knowledge of the context for party cooperation was lacking. Most of
the projects were small and dispersed over many countries. A
recurrent problem was communication and continuity. Institutional
ownership was low; personal ownership high. Most projects involved
training seminars and exchange visits; as other IPA projects often use.
But there appeared to be no follow-up of such events. The evaluation
report concluded that “…..there is little evidence to suggest that NDS
projects have had major lasting impacts on democratic development in
partner organizations, not to mention at a societal level, in the
countries where they have been implemented”(NORAD, 2010:25).
But, as the authors of the report pointed out; that is also the case with
most other types of IPA and can hardly be used as an argument
against the NDS. Seven years would in any case probably be too short
a period to expect to see major changes. The time perspective is
mentioned in almost all evaluation reports and documents on IPA.
However, important reasons for failure in the case of NDS were at the
50
A new evaluation reports has just been published (Braathen & Holm-Hansen, 2014) but
too late to incorporate the findings in this report.
83
organizational level and “….its failure as an arena for learning and
knowledge generation” (NORAD, 2010:25).
Ability to adjust IPA programs as the actors gain experience is a
point raised by several evaluation reports.
US institutions
There are two US institutions engaged in IPA; the International
Republican Institute and the National Democratic Institute which
were established in 1983, when the National Endowment for
Democracy was organized (NRC, 2007). Both institutes work with
several parties, which is a requirement by USAID, one of the major
financial contributors (NRC, 2007). The other main financial
contributor is NED, but other American and international institutions
also support their projects.
The American IPA actors have been active longer than most IPA
donor organizations and have had available considerable financial
resources. It is therefore highly relevant to look at how they see their
own experiences and what evaluation reports tell us about their work.
The two American party affiliations, The National Democratic
The International Republican Institute (IRI)´s mission is advancing
“…freedom and democracy worldwide by helping political parties to
become more issue-based and responsive, assisting citizens to
participate in government planning, and working to increase the role
of marginalized groups in the political process – including women and
youth”51
IRI has had programs in more than 100 countries. Current budget
figures are hard to collect, but in 2008 its budget was about $ 78
mill.52
The National Democratic Institute (NDI)’s long-term goal “..is to
foster a competitive multiparty system by building a range of parties’
abilities to effectively campaign, represent the public, and develop and
implement effective policies (NDI, 2008a: 7). Like its counterpart,
IRI, NDI’s activities concern democracy assistance in general, where
51
http://www.iri.org/learn-more-about-iri/mission
http://rightweb.irc-online.org/profile/International_Republican_Institute#_edn15.
(Accessed August 20 2014)
52
84
support for political parties is one of several types of areas. Its longterm goal for assistance to political parties “is to foster a competitive
multiparty system, rather than advance a particular ideological
position» (NDI, 2008a:4). NDI has since its founding in 1983 been
involved in more than 132 countries.
Evaluation of the American IPA programs have in part been carried
out by USAID, one of the major funders of IPA projects, party by the
donor organizations themselves.
Evaluation of US support in 2007
The USAID report from 2007 (USAID, 2007) assessed the work of
the American efforts to improve political parties in new democracies
and in particularly examined the study of party assistance to Georgia,
Romania, Serbia and Kyrgyzstan. Party assistance is a part of
American foreign aid in general and a component of its democratic
governance program. Assistance focuses on three areas: enhancing
electoral competitiveness, improving internal party organization and
helping parties in governance. Each of these areas has several
components. Electoral competitiveness includes campaigning skills,
media relations, voter mobilization and fundraising. Party
organizational improvement include, among other aspects, building an
extensive organizational network, the development of party statutes,
platform development, membership recruitment, leadership selection
and internal democracy, while governance capacity covers legislative
work, inter-party relationships, communication with the public and
involvement in national, regional and local governments. The report
does not contain an overview of the total financial resources that have
been available for the institutes for the purposes, nor does it cover the
total number of countries or the total time period the institutes have
been engaged in the various countries.
The report notes the academic literature’s findings of variations in
how successful international party assistance has been. Three variables
seem to affect the level of success: 1) Structural factors: the
environment in which party assistance takes place (geography, history,
economic development political culture, political institutions), 2)
Strategy: the approach chosen to support parties, and 3) The role of
the Actors: the donor as well as the recipients (USAID, 2007:9). The
survey of the literature also indicates that the timing of the various
85
forms of assistance can also be critical for its success or not. In the
early phase after the transition to a democratic form of government it
is particularly important to assist with institutional engineering; the
creation of a legitimate electoral process and a fair judicial framework,
rather than on focusing on political parties per se.
Romania provided the most favorable environment of the four
countries studied in the report. Economic conditions were better
there, Romania was ‘surrounded’ by other countries that had transited
to democracy and had the prospect of becoming an EU member.
Moreover, the inter-war period had included a brief period with a
multi-party system. At the other extreme are countries like
Kyrgyzstan and Georgia which have been worse off than Romania on
all structural factors, with Serbia somewhat between. In the report
party building efforts in Romania and Serbia are seen as more
successful than in the two other countries. In Romania and Serbia the
two institutes have work to improve party organizations, particularly
assisting in the establishment of local and regional party organizations.
These efforts have been crucial in building the capacity of the
opposition parties which later succeeded in winning national elections.
Similar efforts did not succeed in Georgia and Kyrgyzstan. Here,
incumbent elites could not be convinced of the value of building party
organizations. Parties, particularly in Kyrgyzstan are “…..largely
devoid in ideology and meaning” (44). Thus, efforts to increase
participation of women and youth (which are other priorities of US
assistance) became meaningless.
NDI and IRI work with several, but not all, political parties in the
recipient countries. In some cases, particularly in transition cases when
several weaker opposition parties faced an entrenched and dominant
party representing the authoritarian regime, US support was given to
the opposition parties. In later periods, all relevant parties, as long as
they were perceived to support democracy, became eligible for
support. The two institutes establish a local office, staffed by US
citizens, but also often work with – or help establish - a local NGO
involved in democracy promotion. But unless the key domestic actors
can be convinced of the need for strong parties, no international effort
can succeed. The report calls this attitude “..the greatest impediment
to success: the tendency among even once reform-minded elites to
reject democratic rules of the game so as to maintain power” (p.13).
Therefore, in the committee’s view it becomes critical to identify the
elite’s incentive structures in the society where the donors operate.
86
The report notes that assistance programs have not been very
successful in promoting internal party democracy, policy development
or linkages between parties and civil society.
As in other evaluation reports the time perspective is important
when assessing the impact of party assistance. In most cases, it will
take a long time to ensure party developments. Increasing
electioneering capacity of the parties is ‘easier’ than building enduring
party organizations. The report argues that the institutes have been
less successful in developing internal party democracy and in
improving linkages between parties and civil society (p. 44).
Evaluation of NDI-support in 2010: the case of Kenya
Indirectly, NDI’s work in Kenya has been evaluated in (NORAD,
2010), which evaluates Norwegian funding of NDI projects. NORAD
contributed to NDI’s project “Pre-election party strengthening in
Kenya”, which ran from 2006 to 2008. NORAD contributed to the
financing of projects before and after the 2007-elections. Pre-election
efforts were directed to building coalitions between opposition
parties, while post-election projects aimed at strengthening the party
structures at the grass-root level, communication between levels in the
party structure, the nomination process in the parties and
communication and interaction between political parties (NORAD,
2010) (p. 6). While coalition-building efforts did take place with
negotiations between the main parties: “…the message clearly came
across that the main purpose of coalition-building was more to gain
access to power for personal enrichment than to gain acceptance for
key policies” (p. 9). While the post-election effort to strengthen grass
root structure made progress, the task is also so formidable that much
more needed to be done. Inter-party relationships were improved by
the creation of dialogues through province level party committees.
However, these seemed to be operating in isolation as national
politicians were unaware of their existence. In spite of the positive
review of NDI’s activities the basic weakness of parties in Kenya
persisted: personality oriented politics, weak internal structures and
democracy and party fragmentation. Clearly, NDI’s work needed to
be continued.
87
Evaluation of IRI-support in 2010
More recently, IRI published in 2010 a report (IRI, 2010)
summarizing its programs in eight countries; Azerbaijan, Cambodia,
Indonesia, Macedonia, Mongolia, Peru, Slovakia and Ukraine. These
eight countries represent quite diverse cases in terms of
democratization. Indonesia, Mongolia and Slovakia have made
considerable progress. Indonesia went from bad to worse until the end
of the 1990s when it changed markedly towards a more democratic
regime. Particularly after 2004, Indonesia has at times been classified
as ‘free’. Mongolia was ranked at the bottom of the Freedom House
democracy scale in 1990 but has since made a consistent climb into the
‘Free’ category. Slovakia oscillated between ‘Free’ and ‘Partly free’ in
the early years after the transition to democracy, but has since 1998
consistently scored ‘Free’. Peru’s status has improved from ‘Partly
free’ to ‘Free’ in 2000, while Macedonia has been stuck in the ‘Partly
free’ category since the records started in 1992-93. Ukraine’ scores
have mostly oscillated between 2 and 4. Azerbaijan has not made any
improvement at all since ratings started in 1991 and have even
regressed from the meager progress in civil rights that were registered
in the late 1990s Cambodia is also firmly stuck in the ‘Not free’
category.
Are IRI’s efforts to strengthen political parties linked to the
differences in democratic trajectory? IRI’s evaluation report looks at
IPA’s contribution to ‘six principles of successful political party
development: party organization, party identity, message development
and delivery, party competition, governance, and strengthening of
legal system” (IRI, 2010:11). 53
The report relied on interviews with IRI staff at headquarters and
in the field offices as well as with party representatives in the
respective countries. The report is quite blunt about the failure of the
party program in Azerbaijan:”…six years of party program work had
not lead to notable improvement in the political party development”
(IRI, 2010:17). Hence, the program switched towards CSOs. Leaders
in CSO’s could be potential political leaders in the future.
But as regards the six principles against which party assistance
could be ‘measured’ the report is much like many of the other
53
It is not clear from the report how these principles have been operationalized and used in
analyzing the country projects.
88
evaluation reports, for IRI as well as for other donor organizations.
The report mentions examples of successful programs aimed at one or
several of the six principles, but is quite scarce with respect to
documenting what has improved and why it has worked in some
circumstances but not in others. Nor are there clearly operationalized
indicators for each of the principles.
However, it is repeated throughout the report the significance of
leadership acceptance of IPA programs. One of the reasons projects
fail is the resistance among party leaders to implement projects they
see as a threat to their own position. This applied to the opposition
parties as well as to the incumbent party. When IPA projects failed to
yield positive results, IRI re-oriented projects partly to CSOs, as
mentioned above, but also to support independent radio stations and
encouraged reforms of legislation relevant for parties. Support for
independent radio stations have increased the opportunity of
opposition parties to communicate their views to the electorate, bypassing the incumbent party’s control with state media institutions.
Indirectly therefore, such projects have improved the competitiveness
in the party system. Reforms of the electoral law and of the party laws
have likewise contributed to levelling the playing field.
UK: The Westminster Foundation for Democracy
The Westminster Foundation for Democracy (WFD) was established
in 1992, is an organization for the British parties, mainly financed by
the Foreign and Commonwealth Office (50% of total revenues 20122013) and DFID (Department of Foreign International
Development), 42%. Some projects are also supported by EU and
other agencies. The total budget for 2013-2014 was £7.08m (WFD,
2013a).
Support for political parties is one of several programs. WFD
works mainly through the three largest parties, Labour, Conservatives
and Liberals, but also supports projects by the smaller parties in the
UK. The main parties’ activities are primarily, but not exclusively,
oriented towards their sister parties in the recipient countries. Among
the topics for party support are: party building and organization
development,
development
of
party
campaigning
and
communications, message and policy formulation, regional co-
89
operation and supporting the capacity of elected representatives at all
levels in political parties.
In the recent business plan (WFD, 2013a) four outputs are
identified, two of which are directly related to political parties 54:
“Minimum of 10 political parties in countries selected by WFD have
strengthened internal structures and external networks, enabling them
to formulate, communicate and campaign on policy based messages
that offer a genuine choice to citizens, and enhanced strategic focus
and
strengthened
co-ordination,
including
party-to-party,
parliamentary and cross-party work; deepened WFD’s technical
expertise and professionalism; reformed WFD structure and
governance arrangements” (p. 3).
Out of a total of £ 5.21 million allocated to the four output areas,
39% is allocated to party support. Of this sum, 30% is spent in Africa
and 33% in Europe. The parties themselves are allocated £2.1 million,
with ca. 41 % each for Labor and Conservatives, 14% for the Liberals
and the remainder (4%) for the smaller parties. All of WFD’s
programs, including the party projects, are in countries that are
eligible for official development aid (WFD, 2013a:49-50).
Overview and evaluation of UK-support in 2009
Whether or not WFD is successful in its work is somewhat disputed.
A 2009 report provided an overview of the British efforts to support
political parties (Wild & Hudson, 2009). It identified party-to-party
support as the main part of the activities, but also with some crossparty initiatives. According to the report WFD did not have a
systematic procedure for evaluating the projects it funded, for example
“There was no evidence…..of the development of common indicators
or standards for measuring the effectiveness of support for political
parties across countries” (Wild & Hudson, 2009:22).
54
The two other targets are: a) Parliamentarians, including female parliamentarians, in 10
legislatures undertake their key legislative, oversight, financial security and representative
roles and b) Civil society organisations in five countries, and women’s groups in three
countries, engage effectively with parliaments, parties and other stakeholders.
90
This overview was later followed by an evaluation of WFD (GPA,
2010). It identified several organizational challenges, particularly the
division between party-to-party activities vs. another objective,
strengthening parliaments, and the overall objective of strengthening
democracy: “.. the purpose of party support - strictly defined – is not
to show demonstrable improvements in the functioning of democracy.
The parties engage in an overtly political set of activities, designed to
help their ideological counterparts in other countries.” Moreover, as is
the case with many evaluation reports, also this report concluded that
“.. we could find no sources which demonstrated the long-term impact
and effect of political party support. Most party-led projects are shortterm, with reports submitted immediately after events; we were not
shown any reports which attempted to gauge the longer term impact
of projects. ……..reports tended to be based on descriptions of
activities that took place, rather than assessing impact against clear
indicators of success” (GPA, 2010: 11-13). 55
Netherlands Institute for Multiparty Democracy
The most comprehensive approach to IPA is probably that organized
by the Netherlands Institute for Multiparty Democracy (NIMD). It
is comprehensive because it includes support for a wide range of
parties and for many different types of activities. NIMD originated as
a common institution for the six major parties represented in the
Dutch parliament. It derives almost all its funding from the Dutch
foreign ministry. Its uniqueness among IPA donors is that it is the
dominant actor in Dutch IPA, even if some of the individual parties
also have a foundation which include party assistance. While the
British WFD is somewhat similar to NIMD at home, abroad the
WFD-parties operate partly individually.
NIMD has been engaged in party assistance programs since 2002
and is now present in 25 countries. NIMD’s vision has been
formulated as contributing to: “Democratic societies in which the
rule of law is observed and the public good fostered”. The vision is
operationalized in three types of programs: 1) promoting inter-party
55
The lack of systematic evaluation and monitoring has since been corrected. The current
business plan refers to a monitoring and evaluation plan with principles and guidelines
applicable for all programs, but a recently conducted evaluation has not yet been made
available to the public (WFD, 2013a).
91
dialogue, 2) support for institutional development of political parties
and 3) strengthening party – civil society relationship. Of the three
components for NIMD’s objectives, the third – strengthening party
civil society partnerships – has been the least developed. The two
other objectives are strongly connected.
NIMD has also the most comprehensive system for evaluation, not
only of its own organization, but also for each of the countries in
which NIMD has been operating 56.
The model that NIMD operates with is that individual political
parties can apply for funds only if they participate in the inter-party
dialogue. In general, NIMD supports parties that succeed in winning
at least one seat in parliament. These parties are invited to establish an
inter-party forum, usually established as an NGO, as for example the
Center for Multiparty Democracy in Malawi (CMD-M). CMD-M is
organized with a board consisting of all parliamentary parties having
two board members, regardless of the size of the party. The
chairmanship rotates between the member parties. If a party fails to
win a seat in a parliamentary election, it loose its membership and
therefore also the possibilities for qualifying for individual funds.
Funding from NIMD is provided to the CMD for running the
secretariat and for funding cross-party activities. The secretariat is
staffed by locally recruited individuals. Each of the member parties
may apply to NIMD centrally for funds to improve their own
organization. Such projects are in principle embedded in a party’s
strategic plan for development where priorities have been established
by the parties’ themselves. Although NIMD centrally decides on the
total funds available for both cross-party and individual projects, the
CMD arrives on an agreement on the formula for how the funds
should be divided. An example is that all parties can receive a basic
fund and additional funds are distributed according the parties’ share
of the seats in parliament.
There are several important distinction from the sister part model.
In the NIMD model all parties with parliamentary representation
qualify for support as long as they participate in the inter-party
dialogue. Thus, the incumbent party is treated in the same way as
opposition parties. The idea is that NIMD should not be seen as an
actor taking sides in the internal politics of the country and that
56
All evaluation reports can be found at www.nimd.org
92
dialogue has priority over individual party projects. NIMD’s financial
support for parties is not to be used for electioneering purposes.
Usually, NIMD’s annual support ceases during election years, or is
greatly reduced during those years. NIMD’s support is also ‘policy
blind’ in the sense that it does not matter which political orientation
recipient parties have – as long as they are committed to upholding a
democratic political system.
CMD’s have been created in several of the countries in which
NIMD is, or has been, operating: Malawi, Mali, Tanzania, Uganda,
Kenya, Ghana, Honduras, and Guatemala. But the local context
influences how NIMD operates. For a CMD to function there must
be a reasonably functioning multiparty system. Thus, in Mozambique
where FRELIMO has been totally dominating there is no CMD. The
local partner office of NIMD supports FRELIMO and the opposition
parties individually. In Georgia, the situation was somewhat similar,
prior to the 2012 parliamentary election. Because of this, the criteria
for including political parties in funding schemes, party performance
in local elections was added as criteria. The structure of the party
system was so unstable that no center for multiparty democracy could
be established. Therefore, NIMD’s office for the Caucasus engaged
individual parties, including UNM, in developing strategic plans. The
office also sponsors cross-party activities, as for example the
interactive web-site for policy issues 57, and seminars aiming to
improve the situation for women in politics, organized jointly with
other donors.
There are also differences across time in how individual party
projects have been structured. In Tanzania and in Malawi NIMD left
it to the parties themselves to identify priorities by first developing
strategic plans, as in the recent program in Georgia. In Uganda NIMD
has a more targeted objective. Its direct support for parties contributes
mainly to ‘policy development’. All participating parties could apply
for funds to engage a policy coordinator. The motivation for
supporting this specific party activity is twofold. NIMD’s strategic
plan states that through improvement of the policy function also
other weaknesses of the parties can be improved: «As indicated in the
Outcome definition, the NIMD program aims specifically to improve the
policy function of parties, which also contributes to the weak performance
57
See http://www.prezidenti.ge/index.php?lang=eng for NIMD’s project in Georgia in
assisting parties and candidates to formulate statements on different policy areas
93
of parties on other objectives”(NIMD, 2012:12-13). Through improved
policy development the electorate will have a clearer choice between
political alternatives.
Thus, NIMD has a long track-record of IPA and has accumulated
considerable knowledge of implementation of programs. Several of
NIMD’s staff members have worked continuously in the organization
for many years and accumulated in depth country expertise.
Evaluations of Dutch support – three country cases
Evaluation reports show that also among NIMD’s projects not all can
be seen as successful, in the sense of improving the party system,
improving party institutionalization and improving the relationships
between civil-society and political parties. The following are examples
of cases (Tanzania, Mali and Ghana) which demonstrate how varied
the results of IPA can be, even when implemented by the same actor
(NIMD).
Tanzania: In Tanzania, NIMD was active from 2002 until 2012.The
evaluation report by Whitehead and Killian (Whitehead & Killian,
2012) provides a generally very positive view of the activities of the
forum for interparty dialogue TCD (Tanzania Center for
Democracy). TCD provided an opportunity for parties to meet and
facilitated debates during elections. Inter party dialogue was perceived
to be so successful that it was recommended also to be used more
systematically at regional and lower levels. But three problems were
noted:
Participation in TCD may have been primarily motivated because it
is a condition for direct party support. Nevertheless, there may have
been positive consequences in terms of inter-party communication.
Opposition parties felt that the incumbent CCM did not take
TCD so seriously, indicated by CCM sending lower level officials to
TCD meetings and sometime also not participating at all in seminars
and meetings where all parties were supposed to participate, In spite of
the dialogue among parties at the top, the prevalence of violence in the
2010 election indicates that there is a limited trickle down-effect inside
the parties.
The authors also noted that the impact of parties participating in
TCD has been quite limited on two key objectives of the programs:
94
internal party democracy and policy based election campaigns.
Election campaigns are still dominated by general promises and on
personalities. Also bilateral partnerships have made little progress in
this respect. Bilateral programs have first of all been used for various
training seminars, but also for more specific organizational purposes,
such as developing strategic plans, developing party constitutions and
printing party documents (p 30). The two main opposition parties,
CHADEMA and CUF, emphasized the positive local “ownership” of
the bilateral programs. Nevertheless, lack of internal democracy was
reported as a problem in all parties ((Whitehead & Killian, 2012:33),
but also that it is very difficult to measure internal party democracy.
Civil society linkage was problematic to evaluate as it is expressly
forbidden in Tanzania for civil society organizations to engage with
political parties. Not much can be said about the impact of TCD in
this respect. In general, civil society organizations do not trust
political parties.
Thus, among the three objectives inter-party dialogue was seen as
successful, strengthening party institutionalization less so, and civil
society-party linkages difficult to handle, at least due to the national
legislation.
Mali: In the case of Mali the program, evaluated by Loada and
Bartholomeussen (Loada & Bartolemeeussen, 2009) had three
components: inter-party dialogue, regional cooperation and bilateral
support for the individual parties (Loada & Bartolemeeussen,
2009:10). The latter aimed at:
•
“strengthening the functioning and internal democracy of the
parties,
•
their capacity to manage conflicts and financial aspects,
•
to strengthen the rules governing how the parties function
etc.”
Although there in the first years were some positive results of the
financing of political parties, the report nevertheless concluded that
many severe problems remained. One of them is the lack of skilled
personnel both in the headquarters and local structures of the parties.
The parties themselves reported on progress on party building efforts,
but the report noted that party organizations were internally weak,
dominated by personalization of politics and lack of institutional
developments. However, as of 2008 the development of strategic plans
95
for parties (and parliamentary groups) has professionalized the parties
and made management of the organizations more efficient: “Many
policymakers have said that the NIMD’s actions have helped improve
the capacities of the parties to commit and to act, and have cited
various arguments in support of such statements” (Loada &
Bartolemeeussen,
2009:29).
Improvements
included
better
preparations for and conduct of election campaigns. Nevertheless,
many (smaller) parties have failed to improve and parties are held in
low regards by the public. The improvement that can be seen in the
individual parties is to some extent due to the improvement in the
running of the center for multiparty democracy (CMDID). At the
same time, the report cautions against believing parties’ ability to
change and adapt should be taken for granted.
The improvement of CMDID has gone hand in hand with
improved management of the larger political parties. It seems that it is
the development of and close monitoring and following up of the
strategic plans that had this ‘transformative’ effect at the level of party
organizations, even if there is still more to be done with respect to
parties’ relationship with the electorate.
A very interesting aspect of the Mali-evaluation report is the
attempt to find explanatory factors for the positive developments.
The report identifies the presidential initiative to organize the Diawara
Commission 58 “forced the parties and the CMDID to commit
themselves in a process of reflecting on multi-party democracy, which
has repercussions on their capacities” (Loada & Bartolemeeussen,
2009:32). NIMD’s support enabled the parties to respond to the
initiative. However, ‘joining’ the presidential consensus in some ways
blurred the distinctive roles of opposition vs. government parties.
Parties that later took a more oppositional role found themselves with
fewer access to governmental resources. The blurring of the
government-opposition boundary is also seen as an explanatory factor
for the increase in the number of independent candidates. The feeling
of local ownership of CMDID and support from the public
authorities has improved the performance of CMDID, as has the
agreement on a set of criteria for party support. The report is clear
about the importance of NIMD’s program for what has been achieved,
regarding CMDID as well as the political parties.
58
A committee appointed by the President for the study of consolidation of democracy.
96
Comparing the Tanzania and the Mali report reveals some
interesting relationship between direct party support and inter-party
dialogue. In the Tanzanian case, the TCD is generally perceived to be
functioning well, but direct support not so well. In the case of Mali
the two go together: improvements in CMDID did lead to positive
developments for the parties.
The two countries are similar with respect to NIMD’s third
objective: strengthening the parties’ relationship with civil society. In
both Mali and Tanzania civil society organizations (CSO) tend not to
trust parties, and CSO’s are expressly forbidden to be linked to
political parties. Even though other components of the Mali program
is evaluated positively the report notes that the electorate hold the
parties in low regards, similar to the situation in Tanzania. However,
we should bear in mind that this is the case throughout many
countries, and may have little to do with the NIMD program as such.
Ghana: Ghana is generally considered to be one of the most
successful examples of democratic consolidation in Sub-Saharan
Africa. The electoral process has gradually become well
institutionalized and managed and when the incumbent party has lost,
it has accepted the outcome, even in very closely fought elections and
there has been a peaceful handover of governmental power. NIMD has
been active in Ghana since 2001 (Dijkstra & Kumado, 2004). As in
most other countries there has been a forum for interparty dialogue,
combined with direct support for individual parties. In Ghana, the
four main political parties committed themselves to a country
program that focused on parties’ role in consolidating constitutional
democracy. A long term strategy was formulated by the parties –a
Joint Action Plan- to achieve this aim. Next to various cross-party
activities, the program included a bilateral support component that
was focused on the strengthening of parties’ organizational and
institutional capacity and linked to the objectives outlined in the Joint
Action Plan (NIMD, 2004a:14-15). The bilateral support component
of the program was thus clearly linked to and guided by the multiparty
dialogue process.
NIMD’s activities in Ghana have run in parallel to, if not caused,
the democratic progress in the country and the Ghanian case is seen as
an example to be followed for parties in other African countries. There
are of course many explanations for why Ghana has become a
successful case of democratic progress, and some scholars, argue that
democracy assistance has been of marginal importance (E.Gyimah97
Boadi & Yaka, 2013). In any case, even if there has been an impact of
democracy assistance, Ghana’s case also illustrates what scholars of
democratization processes have argued are the most important: the
socio-economic and institutional contexts. On these factors Ghana is
scoring better than most other African countries on, for example, the
Human Development Index (HDI) (UNDP, 2014) ; a summary
indicator of socio-economic conditions. Ghana’s parliament, also
considered to be a favorable factor for party building, has more power
than some other African countries where NIMD is active (Fish &
Kroenig, 2009). 59
In contrast to the Swedish model where the PAO chose in which
countries to work, NIMD’s work is strongly linked to official Dutch
development policy. Tanzania and Malawi both received substantial
IPA until 2012 via NIMD, but the two countries are no longer part of
the Dutch government’s priorities. Therefore, NIMD has withdrawn
from the two countries but the local CMD’s continue. 60
59
The Kenyan legislature has been further strengthened as a result of the constitution
enacted in 2010.
60
In Tanzania the Danish Institute for Parties and Democracy has replaced NIMD and in
the case of Malawi, UNDP and CMD-M together are seeking new funding partners.
98
References
Aceron, J. (2009). It's the (non-)system, stupid: Explaining 'maldevelopment' of parties in the Phillipines. In Re-forming the Philippine
political party system: Ideas and initiatives, debates and dynamics (pp. 522): Friedrich Ebert Stiftung.
Agne, H. (2014). Is successful democracy promotion possible? The
conceptual problem. Democratization, 21(1), 49-71.
Aldrich, J. H. (1995). Why Parties. The origin and transformation
of party politics in America. London: University of Chicago Press.
Ames, B., & Power, T. J. (2010). Parties and governability in
Brazil. In P. Webb & S. White (Eds.), Party politics in new democracies
(pp. 179-212). Cambridge: Cambridge University Press.
Assarson, J. (1993). Bør partierna vara internt demokratiska?
Statsvetenskaplig Tidsskrift, 96(1), 39-68.
Bader, M. (2010). Party politics in Georgia and Ukraine and the
failure of Western assistance. Democratization, 17(6), 1085-1107.
Bardi, L., Bressanelli, E., Calossi, E., Gagatek, W., Mair, P., &
Pizzimenti, E. (2010). How to create a transnational party system.
Bruxelles: European Parliament. Directorate-General for Internal
Policies. Policy Department C: Citizens' rights and constitutional
affairs.
Basedau, M., & Stroh, A. (2008). Measuring party
institutionalization in developing countries: A new research
instrument applied to 28 African political parties. Hamburg: GIGA
Working Paper no. 69.
Berge, B. v. d., Poguntke, T., Obert, P., & Tipei, D. (2013).
Measuring intra-party democracy. A guide for the content analysis of
party statutes with examples from Hungary, Slovakia and Romania.
London: Springer
Bergman, T., Muller, W. C., & Strøm, K. (2000). Introduction:
Parliamentary democracy and the chain of delegation. European
Journal of Political Research, 37.
Bertoa, F. C. (2014). Political parties or party systems? Assessing
the "myth" of institutionalization and democracy, from
99
http://whogoverns.eu/wp-content/uploads/2014/10/EPSA2014.pdf
Biezen, I. v. (2003). The place of parties in contemporary
democracies. West European Politics, 26(3), 171-184.
Biezen, I. v. (2005). On the theory and practice of party formation
and adaptation in new democracies. European Journal of Political
Research, 44, 147-174.
Bogaards, M. (2004). Counting parties and identifying dominant
party systems in Africa. European Journal of Political Research, 43,
173-194.
Bolleyer, N., & Storm, L. (2010). Problems of party assistance in
hybrid regimes: the case of Morocco. Democratization, 17(6), 12021224.
Braathen, E., & Holm-Hansen, J. (2014). Review of Norwegian
democracy support via political parties. Oslo: Norad
Brinks, D., & Coppedge, M. (2006). Diffusion is no illusion.
Neighbour emulation in the Third wave of democracy. Comparative
Political Studies, 39(4), 463-489.
Brown, S. (2011). 'Well, what can you expect?': donor officials'
apologetics for hybrid regimes in Africa Democratization, 18(2), 512534.
Burnell, P. (2009). Legislative strengthening meets party support in
international assistance: A closer relationship? Journal of Legislative
Studies, 15(4), 460-480.
Burnell, P. (Ed.). (2006). Globalising democracy. Party politics in
emerging democracies. London: Routledge.
Burnell, P., & Gerrits, A. (2010). Special issue on international
party assistance. Democratization, 17(6).
Burnell, P., & Gerritts, A. (2010). Promoting party politics in
emerging democracies. Democratization, 17(6), 1065-1084.
Carothers, T. (2005). Political party aid: Issues for reflection and
discussion (No. Art.no. SIDA4713en). Stockholm: SIDA.
Carothers, T. (2006). Confronting the Weakest Link. Aiding
Political Parties in New Democracies. Washington, D.C.: Carnegie
Endowment for International Peace.
100
Colomer, J. M. (2005). It's parties that choose electoral systems
(or, Duverger's laws upside down). Political Studies, 53, 1-21.
Cornell, A. (2013). Institutional impediments and reluctant actors.
The limited role of democracy aid in democratic development.
University of Gothenburg, Gothenburg.
Cranenburgh, O. v. (2011). Democracy promotion in Africa: the
institutional context. Democratization, 18(2), 433-461.
Dahl, R. A. (1971). Polyarchy: Participation and opposition. New
Haven, Conn.: Yale University Press.
Dalton, R. J. (2008). The quantity and the quality of party systems.
Party system polarization, its measurements, and its consequences.
Comparative Political Studies, 41(7), 899-920.
Dalton, R. J., Cain, B. E., & Scarrow, S. E. (2006). Democratic
politics and democratic institutions. In B. E. Cain, R. J. Dalton & S. E.
Scarrow (Eds.), Democracy transformed? Expanding political
opportunities in advanced industrial democracies (pp. 250-275).
Oxford: Oxford University Press.
Dalton, R. J., & Wattenberg, M. P. (2002). Unthinkable
democracy. Political change in advanced industrial democracies. In R.
Dalton & M.Wattenberg (Eds.), in Parties without partisans. Political
change in Advanced Industrial Democracies, (pp. 3-18). Oxford:
Oxford University Press.
Dalton, R. J., & Weldon, S. A. (2005). Public images of political
parties. A necessary evil? West European Politics, 28(5), 931-951.
Darnolf, S. (2011). International election support: Helping or
hindering democraic elections. Representation, 47(4), 361-382.
Dijkstra, J. P., & Kumado, K. (2004). REPORT ON THE
EVALUATION OF THE IMD/IEA PROGRAMME IN GHANA
2000 - 2003. the Hague: NIMD.
DIPD. (2013). ANNUAL REPORT 2013 Copenhagen: DANISH
INSTITUTE FOR PARTIES AND DEMOCRACY.
Doherty, I. (2001). Democracy out of balance. Civil society can't
replace political parties. Policy Review, 25-35.
Domingo, P., & et.al. (2012). Joint evaluation: UN
Women/UNDP support to women’s political participation in sub-
101
Saharan Africa. Evaluation synthesis report Commissioned by UN
Women/UNDP. London: ODI.
Duverger, M. (1967). Political Parties. London: Methuen.
E.Gyimah-Boadi, & Yaka, T. (2013). Ghana: The limits of external
democracy assistance. In D. Resnick & N. v. d. Walle (Eds.),
Democratic trajectories in Africa. Unravelling the impact of foreign aid
(pp. 256-280). Oxford: Oxford University Press.
Edwards, M., & Hulme, D. (1996). Too Close for Comfort? The
Impact of Official Aid on Nongovernmental Organizations. World
Development, 24(6), 961-973.
Erdmann, G. (2006). Hesitant befellow: the German Stiftungen and
party aid in Africa. In P. Burnell (Ed.), Globalising democracy. Party
politics in emerging democracies (pp. 181-199). London: Routledge.
Fish, M. S., & Kroenig, M. (2009). The Handbook of National
Legislatures. A Global Survey. Cambridge: Cambridge University
Press.
Fritz, V., & Menocal, A. R. (2007). Developmental States in the
New Millennium: Concepts and Challenges for a New Aid Agenda.
Development Policy Review, 25(5), 531-552.
Giliomee, H., & Simkins, C. (1999). The dominant party regimes
of South Africa, Mexico, Taiwan and Malaysia: A comparative
assessment. In H. Giliomee & C. Simkins (Eds.), The awkward
embrace. One-party dominance and democracy (pp. 1-46). Amsterdam:
Harwood Academic Publishers.
Glenn, J. K. (2000). Civil society transformed: International aid to
new political parties in the Czech Republic and Slovakia. Voluntas,
11(2), 161-179.
Gonzalez-Acosta, E. (2009). Political parties and policy
development. The conditions which lead political parties to adopt
progressive policies. Oslo: UNDP Oslo Governance Centre.
Gouzou, J., Sanou, B. W., & Vedsted, S. (2014). Evaluation of
National Democratic Institute'sNDI) programme 2009-2014 in
Burkina Faso: "Enhancing the ability of women leaders to engage
effectively in pro-poor democratic governance". Stockholm: Sida.
GPA. (2010). Review of the Westminster Foundation for Democracy.
London: Global Partners and Associates.
102
Hahn-Fuhr, I., & Worschech, S. (2014). External democracy
promotion and divided civil society - the missing link. In T. Beichelt,
I. Hahn & F. Schimmelfennig (Eds.), Challenges to Democracy in the
21st Century : Civil Society and Democracy Promotion (pp. 11-41).
New York, N.Y.: Palgrave Macmillan.
Hale, H. E. (2007). Why not parties in Russia? Democracy,
federalism, and the state. Cambridge: Cambridge University Press.
Hearn, J. (1999). Foreign aid, democratization and civil society in
Africa: a study of South Africa, Ghana and Uganda (No. 368).
Brighton: Institute of Development Studies, University of Sussex.
Hellmann, O. (2011). A historical institutionalist approach to
political party organization: The case of South Korea. Government and
Opposition, 46(4), 464-484.
Hicken, A. (2009). Building party systems in developing democracies.
Cambridge: Cambridge University Press.
Hlousek, V., & Kopecek, L. (2010). Origin, ideologi and
transformation of political parties. East-Central and Western Europe
compared. Fornham: Ashgate.
Hudson, A., & Wren, C. (2007). Parliamentary strengthening in
developing countries. London: ODI.
Hulsey, J. W. (2010). 'Why did they vote for those guys again?'
Challenges and contradictions in the promotion of political
moderation in post-war Bosnia and Herzegovina. Democratization,
17(6), 1132-1152.
Huntington, S. P. (1991). The third wave. Democratization in the
late twentieth century. London: University of Oklahoma Press
Hyden, G. (2006). African politics in comparative perspective.
Cambridge: Cambridge University Press.
IIDEA. (2014a). Funding of Political Parties and Election
Campaigns. A Handbook on Political Finance. Stockholm:
International Institute for Democracy and Electoral Assistance.
IIDEA. (2014b). Politics meets policies. The emergence of
programmatic parties. Stockholm: International Institute for
Democracy Assistance.
103
IRI. (2010). A comparative analysis of political party programs at
the International Republican Institute. Studying twenty years of party
development efforts in eight countries. Washington, DC:
International Republican Institute.
Irvine, J. A. (2013). Electoral breakthroughs in Croatia and Serbia:
Women's organizing and international assistance. Communist and
Post-Communist Studies, 46, 243-254.
Kandawasvika-Nhundu, R. (2013). Political Parties in Africa
through a Gender Lens. Stockholm: IDEA.
Katz, R., & Mair, P. (1995). Changing models of party
organization and party democracy: The emergence of the cartel party.
Party Politics, 1(1), 5-28.
Katz, R. S., & Kolodny, R. (1994). Party organizations as empty
vessel: Parties in American politics. In R. S. Katz & P. Mair (Eds.),
How parties organize. Change and adaptation in party organizations in
Western democracies (pp. 23-50). London: Sage.
Keefer, P. (2005). Democratization and clientelism: Why are young
democracies badly governed? Washington DC: The World Bank
Development Research Group.
Keefer, P. (2011). Collective action, political parties and prodevelopment public policy. Washington DC: The World Bank
Development Research Group.
Keefer, P. (2013). Organizing for prosperity. Collective action,
political parties and the political economy of development.
Washington DC: The World Bank Development Research Group.
Kelly, J. G. (2012). International influence on elections in new
multiparty states. Annual Review of Political Science, 15(203-220).
Kjellström, E.-M., Makara, S., & Sjöberg, P. (2009). Party
cooperation in a results perspective. Country study: Uganda. Stockholm:
SADEV.
Kumar, K. (2006). Promoting Independent Media: Strategies for
Democracy Assistance. London: Lynne Rienner.
Kuncic, A. (2011). Aid us to win the elections: Foreign aid and
voter turnout. Developing Economies, 49(3), 233-265.
104
Lavenex, S., & Schimmelfennig, F. (2011a). EU democracy
promotion in the neighbourhood: from leverage to governance?
Democratization, 18(4), 885-909.
Lavenex, S., & Schimmelfennig, F. (2011b). Special Issue:
Democracy promotion in the EU's neighbourhood: From leverage to
governance? Democratization, 18(4).
Lean, S. F. (2007). Democracy assistance to domestic election
monitoring organizations: Conditions for success. Democratization,
14(2), 289-312.
Levitsky, S., & Way, L. (2005). International linkage and
democratization. Journal of democracy, 16(3), 20-34.
Levitsky, S., & Way, L. A. (2010). Competitive authoritarianism.
Hybrid regimes after the cold war. Cambridge: Cambridge University
Press.
Lewis, P. G. (2000). Political parties in Post-Communist Eastern
Europe. London: Routledge.
Lijphart, A. (1994). Electoral systems and party systems. A study
of twenty-seven democracies, 1945-1990. Oxford: Oxford University
Press.
Linz, J., & Stepan, A. (1996). Problems of democratic transititon and
consolidation. Baltimore: Johns Hopkins University Press.
Lipset, S. M. (2000). The indispensability of political parties.
Journal of democracy, 11(1), 48-55.
Lipset, S. M., & Rokkan, S. (1967). Cleavage Structures, Party
Systems and Voter Alignments: An Introduction. In S. M. Lipset & S.
Rokkan (Eds.), Party Systems and Voter Alignments (pp. 1-64). N.Y.:
Free Press.
Loada, A., & Bartolemeeussen, S. (2009). Evaluation of Dutch
support to capacity development. Evidence-based case study of
NIMD's country programme in Mali and of the WARP ACE Europe.
Lupu, N. (2014). Brand dilution and the breakdown of political
parties in Latin America. World Politics, 66(4), 561-602.
Mainwaring, S. (2012). From representative democracy to
participatory competitive authoritarianism: Hugo Chavez and
Venezuelan politics. Perspectives on Politics, 10(4), 955-968.
105
Manning, C. (2010). Mozambique's slide into one-party rule.
Journal of democracy, 21(2), 151-165.
McFaul, M. (2001). Explaining party formation and nonformation
in Russia. Actors, institutions, and chance. Comparative Political
Studies, 34(10), 1159-1187.
MFA. (2013). Final Report. Review of the Danish Institute for
Parties and Democracy. Copenhagen: Ministry of Foreign Affairs.
Michels, R. (1965). Political Parties. N.Y.: Free Press.
Mitchell, P. (2000). Voters and their representatives: Electoral
institutions and delegation in parliamentary democracies. European
Journal of Political Research, 37, 335-351.
Moehler, D. C. (2010). Democracy, governance, and randomized
development assistance. Annals of the American Academy of Political
and Social Science(668), 30-46.
Morgan, J. (2011). Bankrupt representation and party system
collapse. University Park, Pa.: Pennsylvania State University Press.
Muller, W. C. (2000). Political parties in parliamentary
democracies: Making delegation and accountability work. European
Journal of Political Research, 37, 309-333.
Muriaas, R. L., Tønnessen, L., & Wang, V. (2013). Exploring the
relationship between democratization and quota policies in Africa.
Women's Studies International Forum, 41(1), 89-93.
Myers, D. J. (1990). Venezuela: The politics of liberty, justice and
distribution. In H. Wiarda & H. F. Kline (Eds.), Latin American
politics and development (pp. 283-320). Boulder, Co.: Westview Press.
NDI. (2008a). A Guide to Political Party
Washington, DC: National Democratic Institute.
Development.
NDI. (2008b). Minimum standards for the democratic functioning
of political parties. Washington, DC: National Democratic Institute.
Nenadović, M. (2010). An uneasy symbiosis: the impact of
international administrations on political parties in postconflict
countries. Democratization, 17(6), 1153-1175.
NIMD. (2003). Annual report 2002. The Hague: NIMD.
NIMD. (2004a). Annual report 2003. The Hague: NIMD.
106
NIMD. (2004b). A framework for democratic party-building. den
Haag: NIMD.
NIMD. (2007). Political parties: Pillars of democracy. Multi-annual
plan 2007-2010. den Haag: NIMD.
NIMD. (2011). Annual report 2010. den Haag: NIMD.
NIMD. (2012). Multi Annual Plan. den Haag: NIMD.
NIMD. (2014a). Annual report 2013. The Hague: NIMD.
NIMD. (2014b). NIMD Annual Report 2013. The Hague: NIMD.
NORAD. (2010). Evaluation of the Norwegian Centre for
Democracy Support 2002-2009. Oslo: NORAD.
NRC. (2007). Improving democracy assistance. Building
knowledge through evaluations and research. Washington, D.C.:
National Research Council.
OECD. (2012). Closing the Gender Gap: Act Now. Paris: OECD.
Ostrogorski, M. (1964). Democracy and the organization of political
parties (Vol. II). N.Y.: Anchor Books.
Pasquino, G. (2006). Studying the never ending Italian transition.
European Political Science, 5, 423-433.
Peck, L., Kjellström, E.-M., Kuzio, T., & Sjöberg, P. (2009). Party
cooperation in a results perspective. Country study: Ukraine. Stockholm:
SADEV.
Peck, L., Rudqvist, A., & Ramos, M. J. (2009). Party cooperation
in a results perspective. Country study: Colombia. Stockholm:
SADEV.
Peeler, J. (1998). Building democracy in Latin America. London:
Lynne Rienner.
Poguntke, T., Obert, P., & Tipei, D. (2013). Measuring intra-party
democracy. A guide for the content analysis of party statutes with
examples from Hungary, Slovakia and Romania. New York: Springer.
Powell, N., & Tucker, J. A. (2014). Revisiting electoral volatility in
Post-Communist countries: New data, new results and new
approaches. British Journal of Political Science, 44(1), 123-147.
107
Power, G. (2008). Donor support to parliaments and political
parties: An analysis prepared for DANIDA: Global Partners &
Associates.
Power, G., & Coleman, O. (2011). The Challenges of Political
Programming: International Assistance to Parties and Parliaments.
Stockholm: IDEA.
Pridham, G. (Ed.). (1990). Securing democracy: Political parties
and democratic consolidation in Southern Europe. London:
Routledge.
Przeworski, A., Stokes, S. C., & Manin, B. (Eds.). (1999).
Democracy, accountability and representation. Cambridge: Cambridge
University Press.
Rakner, L., & Svåsand, L. (2010). In search of the impact of
international support for political parties in new democracies: Malawi
and Zambia compared. Democratization, 17(6), 1250-1274.
Rakner, L., & Svåsand, L. (2011). Foreign aid, democracy
promotion and democratic progress. In J. Blom-Hansen, C. GreenPedersen & S.-E. Skaaning (Eds.), Democracy, elections and political
parties (pp. 39-48). Århus: Politica.
Randall, V., & Svåsand, L. (2002). Party institutionalization in new
democracies. Party Politics, 8(1), 5-29.
Regjeringen. (2010). Strategi för särskilt demokratistöd genom
svenska partianknutna organisationer 2011-2015.
Resnieck, D., & Walle, N. v. d. (Eds.). (2013). Democratic
Trajectories in Africa. Unravelling the Impact of Foreign Aid. Oxford:
Oxford University Press.
Risse, T., Magen, A., & McFaul, M. (Eds.). (2009). Promoting
Democracy and the Rule of Law : American and European Strategies.
Basingstoke, Hampshire, GBR: Palgrave Macmillan.
Rodan, G. (2012). Competing ideologies of political representation
in Southeast Asia. Third World Quarterly, 33(2), 311-332.
Rokkan, S. (1970). Nation-Building, Cleavage Formation and the
Structuring of Mass Politics. In S. Rokkan (Ed.), Citizens, Elections,
Parties. Oslo: Universitetsforlaget.
108
Romeo, L. G. (2003). The role of external assistance in supporting
decentralisation reform. Public Administration and Development,
23(1), 89-96.
Rose, R., & Shin, D. C. (2001). Democratization backwards: The
problem of third-wave democracies. British Journal of Political Science,
31(2), 331-354.
Roza, V., Llanos, B., & Roza, G. G. d. l. (2012). Gender and
political parties. Far from parity. Stockholm: IDB/IDEA.
SADEV. (2009). Party cooperation in a results perspective.
Evaluations of support to democracy through Party Affiliated
Organisations (PAO) (No. 2009:3). Stockholm: Swedish Agency for
Development Evaluation.
Samuels, D. J., & Shugart, M. S. (2010). Presidents, parties and
prime ministers. How the separation of powers affects party
organization and behavior. Cambridge: Cambridge University Press.
Sartori, G. (1976). Parties and Party Systems. Cambridge:
Cambridge University Press.
Schattschneider, E. S. (1942). Party government. N.Y.: Holt,
Rinehart and Winston.
Schedler, A. (1999). Conceptualizing accountability. In L.
DIamond, M. Plattner & A. Schedler (Eds.), The self-restraining state:
Power and accountability in new democracies (pp. 13-28). London:
Lynne Rienners.
Schmitter, P. C. (2001). Parties are not what they once where. In L.
Diamond & R. Gunther (Eds.), Political parties and democracy (pp.
67-89). London: Johns Hopkins University Press.
Shugart, M. S. (1998). The inverse relationship between party
strength and executive strength: A theory of politicians' constitutional
choices. British Journal of Political Science, 28(1), 1-29.
Sida. (2014a). Halvtidsöversyn av genomförandet av strategin for
särskilt demokratistöd genom svenska partianknutna organisationer
2011-2015. Stockholm: SIDA.
Sida. (2014b). PAO organisationer, stöd, program-insatser,
geografisk stödområde.
109
Sikk, A. (2005). How unstable? Volatility and the genuinely new
parties in Eastern Europe. European Journal of Political Research, 44,
391-412.
SIPU. (2011). Anvisning för det särskilda demokratistödet riktat til
politiska partier och flerpartisystem genom svenska partianknutna
organisationer (PAO).
Smith, G. (2009). Democratic innovations. Designing institutions
for citizen participation. Cambridge: Cambridge University Press.
Southall, R. (2001). Opposition in South Africa: Issues and
problems. In R. Southall (Ed.), Opposition in South Africa (pp. 1-24).
London: Frank Cass.
Spoerri, M. (2010). Crossing the line: partisan party assistance in
post-Milosovic Serbia. Democratization, 17(6), 1108-1131.
Stewart, S. (2009). Democracy promotion before and after the
'color revolutions'. Democratization, 16(4).
Stokes, S. C. (1999). Political parties and democracy. Annual
Review of Political Science, 2, 243-267.
Stokes, S. C. (1999). What do policy switches tell us about
democracy? In A. Przeworski, S. C. Stokes & B. Manin (Eds.),
Democracy, accountability, and representation (pp. 98-130).
Cambridge: Cambridge University Press.
Stoner-Weiss, K. (2001). The limited reach of Russia's party
system: Underinstitutionalization in dual transitions. Politics and
Society, 29(3), 385-414.
Sumich, J. (2010). The party and the state: Frelimo and social
stratification in post-socialist Mozambique. Development and Change,
41(4), 679-698.
Svåsand, L. (2013). Party Development in the Old World: And in
the New. In W. C. Muller & H. M. Narud (Eds.), Party Governance
and Party Democracy (pp. 253-274). New York: Springer.
Svåsand, L. (2014). Regulation of political parties and party
functions in Malawi: Incentive structures and the selective application
of the rules. International Political Science Review, 35(3).
Teorell, J. (1999). A deliberative defence of intra-party democracy.
Party Politics, 5(3), 363-382.
110
Thapa, G. B., & Sharma, J. (2009). From insurgency to democracy:
The challenges of peace and democracy building in Nepal.
International Political Science Review, 30(2), 205-219.
Uggla, F., Benneck-Björkman, L., Hadenius, A., Norvalln, F.,
Tamra, A., & Öhman, M. (2000). Stöd till de partianknutna
partiorganisationerna. Stockholm: Department for Cooperation with
Non-Governmental Organisations and Humanitarian Assistance.
UNDP. (2005). UNDP's Engagement with Political Parties. Oslo:
UNDP
Oslo
Governance
Centre
(http://www.undp.org/oslocentre/overview/ogc_political_parties.htm
l).
UNDP. (2006). A Handbook on Working with Political Parties.
Oslo:
UNDP
Oslo
Governance
Center
(http://www.undp.org/oslocentre/overview/ogc_political_parties.htm
l).
UNDP. (2014). Human Development Report 2014. Sustaining
Human Progress: Reducing Vulnerabilities and Building Resilience.
New York: UNDP.
UNDP/NDI. (2012). Empowering Women for Stronger Political
Parties. A Guidebook to Promote Women’s Political Participation.
New York: UNDP/NDI.
USAid. (2007). A study of political party assistance in Eastern Europe
and Eurasia. Washington DC: US Agency for International
Development.
Vinnai, V. (2007). Demokratieförderung in Afrika. Die
Zusammenarbeit der Friedrich-Ebert-Stiftung mit politischen Parteien
und Befreiungsbewegungen in Afrika. Berlin: LIT Verlag.
Waal, T. d. (2012). Georgia
counterrevolution. Bloomberg.
is
having
a
democratic
Wang, C.-H. (2014). The effects of party fractionalization and
party polarization on democracy. Party Politics, 20(5), 687-699.
Weissenbach, K. (2010). Political party assistance in transition: the
German ‘Stiftungen’ in sub-Saharan Africa. Democratization, 17(6),
1225-1249.
111
Weissenbach, K. (2013). Outside support for internal party
democracy? How the German political foundations shape political
parties in Sub-Saharan Africa. Bordeaux: ECPR General Conference.
Weissenbach, K. (2014). “With a little help from my friends”. The
German political foundations between partisan approach and
multiparty approach in the field of party assistance: ECPR General
Conference,Glasgow 3.-6. September.
WFD. (2013a). Westminster Foundation for Democracy. Business
Plan 2013-2014. London: WFD.
WFD. (2013b). WFD Business plan
Westminster Foundation for Democracy.
2013-2014.
London:
Whitehead, R., & Killian, B. (2012). Evalution report.NIMD's
Tanzania Programme- 2007-2011.
Wild, L., & Hudson, A. (2009). UK support for political parties: A
stock take. London: ODI.
Wyman, M., White, S., Miller, B., & Heywood, P. (1995). The
place of 'party' in post-communist Europe. Party Politics, 1(4), 535548.
Xezonakis, G. (2012). Party System Polarisation and Quality of
Government: On the Political Correlates of QoG. Gothenburg: QOG
THE QUALITY OF GOVERNMENT INSTITUTE, Department of
Political Science, University of Gothenburg.
Young, D. J. (2014). An initial look into party switching in Africa:
Evidence from Malawi. Party Politics, 20(1), 105-115.
Öhman, M., Öberg, S. A., Holmström, B., Wockelberg, H., &
Åberg, V. (2004). Politiska partier och demokratibistånd. Översyn av
stödet
genom
svenska
partianknutna
organisationer
till
demokratiuppbygnad i u-lander og och lander i Central- och
Östeuropa. Stockholm: Sida. Department for Democracy and Social
Development.
112